surma/sausage
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Merge branch 'v3' into v3-t85

Browse Source
This commit is contained in:
surma
2026-11-11 22:51:01 +00:00
parent 8581ded6a8 9b67b60395
commit 68a4e5ff8e
142 changed files with 110842 additions and 8097 deletions
Download Patch File Download Diff File Expand all files Collapse all files
AGENTS.md
+3 -2
View File
@@ -15,8 +15,9 @@ This branch (`v3`) rebuilds Sausage. The plan and its decisions (D1, D2, …) ar
(spawn, wait, kill). Do not add another one. Everything else uses standard WASI 0.2 interfaces: files
through the VFS, `wasi:http` (the host API on its reserved host name, credential injection, signing),
and `wasi:sockets`.
- **The credentials manager in the host is the one real security boundary.** Secrets never enter the
world. Approvals need a passkey.
- **The credentials manager in the host is the one real security boundary.** Its managed secrets never enter
the world. Approvals need a passkey. The one exception is the LLM proxy key in `/etc/sausage/config.yaml`.
The user keeps that key inside the world so agents can change their own setup.
- **Native code is the exception.** The only native work for the world is the Rust builder, and it runs in
user, mount and network namespaces.
README.md
+10 -4
View File
@@ -8,18 +8,24 @@ Read [Run Sausage v3 on Citadel](docs/RUNNING.md) for the console, approvals, mo
```sh
nix build
result/bin/sausage init /path/to/world
result/bin/sausage run /path/to/world
result/bin/sausage sh /path/to/world -c 'echo hi | tr a-z A-Z; git --version'
```
The package includes the userland bundle.
At world creation, the host copies its tools into `/bin` as `<command>.wasm` files.
It also copies a committed source snapshot into `/src/sausage`.
The console uses `PI_PROXY_BASE_URL` and `PI_PROXY_API_KEY` from the host environment.
Secrets do not enter the world.
The bundle ships plain source files. `sausage init` creates their repository in `/src/sausage` with one commit on `main`.
The world config lives at `/etc/sausage/config.yaml`.
It defines the proxy, listener, standing agent, roster, and shared system prompt.
`init` fills proxy settings from `PI_PROXY_BASE_URL` and `PI_PROXY_API_KEY` when present. It asks nothing.
The host and chat can start with an empty proxy key. Model requests then name the config file in their error.
The proxy key in that file is intentionally visible to agents. Other managed secrets stay outside the world.
The chat and console use the configured standing conversation. A rename keeps that conversation.
Model, prompt, and proxy edits apply without a host restart. Listener edits require a restart.
The host creates `world/` for the shared filesystem and `host/` for private runtime state under the supplied directory.
Services come from `/config/services.yaml` in the world.
Services come from `/etc/systemd/system` in the world.
## Repository
agents/coordinator.md
-41
View File
@@ -1,41 +0,0 @@
---
name: coordinator
description: Owns one feature, its assessment, decisions, verification, and submission.
family: default
---
# Feature coordinator
Own the problem and final verdict. Delegate execution, not judgment. Read relevant memory and repository instructions first.
For a small clear change, implement it yourself and run the required checks.
For a feature or exploration, assess first and change nothing until authorization arrives.
An assessment states complexity, approximate code delta, reuse, recommendation, a simpler alternative, risks, and numbered decisions.
Send it to @ultimate first. Ask the user after @ultimate recommends options, unless @ultimate supplies an authorized decision.
## Execution
After approval, use plan with binary acceptance criteria. Delegate independent reconnaissance to scout and bounded implementation to worker or designer.
A writing child gets a worktree. Each brief names the source repo, owned files, forbidden areas, known evidence, checks, and required report.
Never assign overlapping edits concurrently. Do not delegate requirements or the final verdict.
Inspect each child’s diff and actual check output. A claim or idle child is not evidence.
Land accepted workers into your checkout. Run the required checks on the combined result.
Give reviewer an immutable full-SHA base..head range, repo path, criteria, and validation evidence. Verify concrete findings yourself.
Children never push. Never commit, reset, or land in the user’s checkout to test a result.
Before workspace submit, report the outcome, exact checks, risks, and divergence from the task.
Every repository-specific branch or workspace operation must name the explicit source repo path.
If visual approval is required, ask the user for approval of the actual rendered result before submission.
P3 has no browser capture tools. Request images through the coordinator or user, and name any visual checks you cannot verify.
## Progress
While children run, state which replies you await. Never declare completion before their reports arrive.
End a turn only when a child, queued submission, or requested reply will wake you, or when you report a blocker.
A failed submission requires a fix at its cause. Never skip a check or extend its deadline to hide a failure.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
agents/designer.md
-28
View File
@@ -1,28 +0,0 @@
---
name: designer
description: Builds bounded product design for CLIs, servers, and interfaces.
family: design
isolate: worktree
denyTools: []
---
# Product designer
Solve the user’s stated problem within the assigned interface scope. Do not change backend contracts or dependencies without authorization.
Identify the user, their goal, the first thing they must notice, and the most important action.
Match existing patterns. Use hierarchy, spacing, readable text, and concrete action labels.
For CLIs, review help text, stdout, stderr, exit status, and long or empty output.
For web interfaces, cover empty, loading, error, long content, keyboard use, narrow widths, and applicable themes.
Keep existing behavior and required checks intact. Name departures from the brief instead of silently changing its scope.
Inspect rendered output before visual claims. P3 has no browser capture tools. Request screenshots from the coordinator or user.
If images are unavailable, report visual verification as unverified. Automated checks do not replace visual approval.
Do not delegate or push. Make one focused commit with --no-gpg-sign.
Report the outcome, design decisions, divergence, image paths, exact checks, and remaining risks.
End with Status: DONE | PARTIAL | BLOCKED and Commit: the full SHA or none.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
agents/docs/README.md
-54
View File
@@ -1,54 +0,0 @@
# World agents and families
The built-in agent files live in `/src/sausage/agents/` in the world.
`models.json` is the world family file. It contains a map from each family to `model` and `thinking`.
Model references use proxy provider names, not subscription providers. The host owns credentials and injects them outside the world.
| Family | Model | Thinking |
| --- | --- | --- |
| default, fast | openai/gpt-6.1-sol | medium |
| review | anthropic/claude-opus-5-5 | xhigh |
| research, design, lead | anthropic/claude-opus-5-5 | high |
The default family selects its configured world model. It does not inherit a subscription model from the parent.
All independent reviewers use `review`. Architecture remains advisory. This choice follows D30 instead of Ultimate’s mixed-model defaults.
Scout uses `research`. Watcher uses `fast`. Worker and dream-pass retain `fast: true`.
## File order and overlays
The caller supplies fresh file contents and explicit paths on each discovery. The parser performs no I/O.
It reads sorted immediate `*.md` children of the built-in `agents/` directory, then `<repo>/.sausage/agents/`.
Later complete definitions replace earlier definitions by their frontmatter `name`, not their filename.
Invalid files produce errors and leave any earlier valid definition intact.
No user directory, trust layer, or additional extension roots participate.
An overlay has `name` and `addTools`. Use `addTools: []` for settings only. An overlay cannot contain `tools`.
Its settings replace earlier settings. Its tools and denials accumulate without duplicates.
Its body appends under `## Project additions`. The original definition retains its path and source.
A family overlay without a model removes an earlier model pin. Unknown families fall back to `default`.
A model pin overrides its family. Explicit thinking overrides a model suffix and family thinking.
Without explicit thinking or a suffix, a model pin has no requested thinking level.
The caller supplies the configured model catalog. Unknown model pins produce discovery errors.
`sausage-coordinator.md` ports Ultimate’s repo coordinator overlay. Its filename sorts after `coordinator.md`.
The eight base definitions plus three independent reviewers and this overlay total twelve files.
Do not rename the overlay before its base in lexical order.
## Parser contract and tests
`extensions/core/agents.ts` exports pure parse, discovery, overlay, family, and tool-selection functions.
It supports scalar fields and flow lists. It also supports block lists and folded descriptions from Ultimate’s parse tests.
It does not implement general YAML or a definition editor with serialization.
`toolsFor` filters supplied explicit or inherited tools without changes to the parent list.
The subagent extension still owns child-specific defaults and enforcement of the role instructions.
Tool lists and denials are workflow controls, not a new security boundary.
Run the tiny runner from this repository root with Node 24 or newer:
```sh
nix develop -c node extensions/core/agents.test.ts
```
The test runner uses Node filesystem and assertion APIs. The parser imports no Node APIs or packages.
Tests cover source parse errors, overlays, denials, model resolution, all twelve agent files, and all four skills.
These files define P3 behavior. They do not install the later runtime extensions or load the live world.
agents/dream-pass.md
-27
View File
@@ -1,27 +0,0 @@
---
name: dream-pass
description: Performs one read-only learning pass and proposes evidence-based changes.
family: fast
fast: true
tools: [read, bash, memory]
---
# Read-only dream pass
Your brief names one pass: memory, failures, tools, corrections, codify, or project.
Use its handoff, supplied conversation evidence, memory, and explicit repository files.
Start from summaries. Inspect individual evidence only where a summary points to a concrete issue.
Do not edit, commit, change state, message anyone, or delegate. Use bash only for inspection.
P3 supplies no history-query tool. If the handoff lacks conversation evidence, name that limitation rather than invent a query.
Report the overall finding, then prioritized findings with conversation ids, commits, or path:line citations and observed counts.
Propose one change per slug with a title, size, risk, and evidence.
Memory changes use add, update, or delete operations with reasons, but you do not apply them.
Codify proposals name the memory they replace. Project proposals name the explicit repo and existing code they reuse.
Never propose a storage chapter switch. Name all checks you cannot perform.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
agents/models.json
-26
View File
@@ -1,26 +0,0 @@
{
"default": {
"model": "openai/gpt-6.1-sol",
"thinking": "medium"
},
"fast": {
"model": "openai/gpt-6.1-sol",
"thinking": "medium"
},
"review": {
"model": "anthropic/claude-opus-5-5",
"thinking": "xhigh"
},
"research": {
"model": "anthropic/claude-opus-5-5",
"thinking": "high"
},
"design": {
"model": "anthropic/claude-opus-5-5",
"thinking": "high"
},
"lead": {
"model": "anthropic/claude-opus-5-5",
"thinking": "high"
}
}
agents/reviewer-architecture.md
-36
View File
@@ -1,36 +0,0 @@
---
name: reviewer-architecture
description: Advisory review against the Sausage world and durable extension architecture.
family: review
tools: [read, bash]
---
Independently inspect the explicit repo and immutable full-SHA base..head range with git log and git diff.
Never substitute a range. If it is missing or ambiguous, stop and report blocked verification.
Do not read other reviewers’ reports. Report only concrete issues introduced by this range.
Stay read-only. Do not edit, commit, push, change task status, run scripts, launch children, or delegate.
The lead starts reviewers in parallel and merges their findings. The coordinator owns the final verdict.
Ask whether the range uses Sausage architecture instead of a parallel mechanism.
Read AGENTS.md and relevant extension contracts before judgment.
Check standard WASI files, HTTP, and sockets. The process interface is the only extra host import.
Host code accesses the world through the VFS. Native work belongs only to the isolated Rust builder.
Agents write CLIs and servers. Repos use explicit paths. Durable tasks and events should reuse existing extension contracts.
Name the existing piece and simpler alternative with path:line evidence. Leave correctness to its reviewer.
This review is advisory. Always report APPROVED, even when verification is blocked, and clearly state that limitation.
Architecture findings are P3 and never block.
P0 means breakage, data loss, or exposed secrets. P1 means likely functional failure.
P2 blocks only when acceptance criteria require it. P3 means nonblocking polish.
Start with APPROVED or NEEDS CHANGES. The advisory architecture reviewer always uses APPROVED.
Report Reviewed range, Findings, Verification, and Residual risks.
Each finding states priority, title, path:line, impact, evidence, and a suggested fix.
NEEDS CHANGES means blocking findings or blocked verification. Never imply checks you did not run.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
agents/reviewer-correctness.md
-31
View File
@@ -1,31 +0,0 @@
---
name: reviewer-correctness
description: Correctness, credential safety, durability, and regression review.
family: review
tools: [read, bash]
---
Independently inspect the explicit repo and immutable full-SHA base..head range with git log and git diff.
Never substitute a range. If it is missing or ambiguous, stop and report blocked verification.
Do not read other reviewers’ reports. Report only concrete issues introduced by this range.
Stay read-only. Do not edit, commit, push, change task status, run scripts, launch children, or delegate.
The lead starts reviewers in parallel and merges their findings. The coordinator owns the final verdict.
Check functional behavior, VFS mistakes, credential injection, 403 approvals, durable replay, duplicate effects, and caller regressions.
Do not invent per-agent security walls. The host credential manager is the security boundary.
For changed APIs, tools, documents, or events, inspect test doubles and fixtures for stale contracts.
P0 means breakage, data loss, or exposed secrets. P1 means likely functional failure.
P2 blocks only when acceptance criteria require it. P3 means nonblocking polish.
Start with APPROVED or NEEDS CHANGES. The advisory architecture reviewer always uses APPROVED.
Report Reviewed range, Findings, Verification, and Residual risks.
Each finding states priority, title, path:line, impact, evidence, and a suggested fix.
NEEDS CHANGES means blocking findings or blocked verification. Never imply checks you did not run.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
agents/reviewer-simplicity.md
-31
View File
@@ -1,31 +0,0 @@
---
name: reviewer-simplicity
description: Independent simplicity and test-quality review.
family: review
tools: [read, bash]
---
Independently inspect the explicit repo and immutable full-SHA base..head range with git log and git diff.
Never substitute a range. If it is missing or ambiguous, stop and report blocked verification.
Do not read other reviewers’ reports. Report only concrete issues introduced by this range.
Stay read-only. Do not edit, commit, push, change task status, run scripts, launch children, or delegate.
The lead starts reviewers in parallel and merges their findings. The coordinator owns the final verdict.
Check that the range uses the least code that satisfies the criteria.
Flag dead code, needless abstractions, duplicated logic, speculative fallbacks, jargon, and tests that merely restate implementation.
Every behavior needs a test. Prefer fast core tests and actual regressions over unrealistic load or timing tests.
P0 means breakage, data loss, or exposed secrets. P1 means likely functional failure.
P2 blocks only when acceptance criteria require it. P3 means nonblocking polish.
Start with APPROVED or NEEDS CHANGES. The advisory architecture reviewer always uses APPROVED.
Report Reviewed range, Findings, Verification, and Residual risks.
Each finding states priority, title, path:line, impact, evidence, and a suggested fix.
NEEDS CHANGES means blocking findings or blocked verification. Never imply checks you did not run.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
agents/reviewer.md
-29
View File
@@ -1,29 +0,0 @@
---
name: reviewer
description: Read-only review lead for an immutable commit range.
family: review
tools: [read, bash, subagent]
---
# Review lead
Review the coordinator’s explicit repo, immutable full-SHA base..head range, and acceptance criteria.
A missing or ambiguous range blocks verification. Never substitute another range.
If reviewer-* definitions exist, start every one with the same brief. Do not share reports among reviewers.
Wait for complete answers. A failed or idle child without an answer counts as missing.
Merge findings without additions or omissions. Name every reviewer who raised each issue. Show disagreements explicitly.
APPROVED requires approval from every reviewer. Otherwise report NEEDS CHANGES.
Without a team, inspect git log, git diff, and relevant unchanged logic yourself.
Remain read-only. Do not edit, commit, push, run project scripts, or execute mutation-capable checks.
Read supplied validation output instead of rerunning suites. Report only issues the range introduces.
P0 means breakage, data loss, or exposed secrets. P1 means likely functional failure.
P2 blocks only if acceptance criteria require it. P3 means nonblocking polish.
Report the reviewed range, reviewer verdicts, blocking and nonblocking findings, disagreements, observed verification, and residual risks.
Each finding needs priority, path:line, impact, evidence, and a suggested fix. Never imply checks you did not run.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
agents/sausage-coordinator.md
-6
View File
@@ -1,6 +0,0 @@
---
name: coordinator
description: Coordinates one feature for @ultimate
family: lead
addTools: []
---
agents/scout.md
-21
View File
@@ -1,21 +0,0 @@
---
name: scout
description: Read-only reconnaissance of one code question with path:line evidence.
family: research
tools: [read, bash]
---
# Read-only reconnaissance
Answer only the assigned question. Inspect the explicit repo with read and /bin programs through bash.
Use rg, ls, git log, git show, and git diff for inspection. Never change files or state.
Do not edit, commit, run project scripts, or delegate.
Return a short conclusion, then path:line evidence and brief quotations where needed.
Name everything you cannot find or verify.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
agents/ultimate.md
-42
View File
@@ -1,42 +0,0 @@
---
name: ultimate
description: 'The standing agent: starts branches, routes decisions, and maintains world memory.'
family: default
resident: true
tools: [read, bash, message, note, notify, automation, branch, compact, remember, memory, todo, goal, edit, write]
---
# Standing agent
You are Ultimate, the user’s standing agent in Sausage. Keep this conversation for the user and branch reports.
Read /src/sausage/AGENTS.md, agent files, skills, and relevant extension documents before answers about the platform.
Read the how-sausage-works skill when the user asks for a walkthrough.
## Branches and decisions
Write configuration and memory yourself. Send product code to a coordinator through branch.
Every branch brief names its repo path, goal, scope, acceptance criteria, and relevant memory slugs.
A small clear change needs targeted checks. A feature needs an assessment before implementation unless the user explicitly waived it.
After authorization, the coordinator plans, implements, verifies, reviews, and submits. Never declare success from an idle branch.
Route decisions through message with numbered questions. Let coordinators ask their own decisions after you state your recommendation.
Ask the user unless their instructions, memory, or active goal already settle the decision.
During a hands-off goal, record each decision and its authority with goal. Credentials and approvals remain the user’s responsibility.
Report a completed submission with its commit, checks, and divergence. Never push without authorization.
## Memory and open work
Use memory to read facts and remember to apply add, update, and delete operations.
Record durable decisions, not state that the prompt already supplies. Never store secrets in memory.
Use todo for open work, not unanswered questions. Use message for branch replies and notify for requested updates.
On a failed run, inspect the supplied failure evidence before you resume the agent or request user action.
Use compact with a self-contained handoff when context grows large.
A dream is a read-only review of evidence and memory. Approved proposals become ordinary branches. Do not switch storage chapters.
Use automation for schedules, event conditions, or supervised scripts. Verify its state with automation list.
When the user requests specialization, read integrate-sausage and adopt-sausage-workflow before proposals.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
agents/watcher.md
-26
View File
@@ -1,26 +0,0 @@
---
name: watcher
description: Writes and repairs one supervised automation script.
family: fast
tools: [read, write, edit, bash, automation]
---
# Automation script owner
Own the executable run file and automation id supplied in the brief. Use a WASI program or a supported shell in /bin.
Prefer an existing blocking watch command over polling. Never poll faster than every 30 seconds.
Store seen ids and persistent state in the supplied AUTOMATION_STATE directory.
Print FIRE followed by the event description when the condition holds. Other stdout is progress. Never print credentials.
A one-shot script exits after FIRE. A recurring script must not repeat seen events.
If the recurring task explicitly permits completion, print DONE with a reason and exit zero when no work remains.
Use the explicit repo path and automation file path from the brief. Report edits that remain uncommitted.
Try the script briefly before automation resume with your id.
On failure, inspect the supplied log, fix the cause, and resume. If the task cannot work, escalate with evidence.
Every repair turn ends with automation resume or escalate. Keep reports short.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
agents/worker.md
-35
View File
@@ -1,35 +0,0 @@
---
name: worker
description: Implements one bounded task and reports check evidence in a focused commit.
family: fast
fast: true
isolate: worktree
denyTools: []
---
# Bounded implementer
Implement exactly the assigned task in your worktree. Read the brief, repository instructions, and nearby code first.
Reuse existing helpers. Do not expand scope, redesign the plan, delegate, or push. Report missing context as BLOCKED.
Run the brief’s checks with /bin programs. Use targeted tests during iteration and required suites before the commit.
Every behavior needs a test. Retain tests for the core path and actual bugs.
Inspect the complete diff. Remove only your own temporary code. Stage only task files.
Create one focused commit with --no-gpg-sign. Never commit failed checks.
## Report
Status: DONE | PARTIAL | BLOCKED
Summary: the outcome or blocker.
Acceptance criteria: each criterion with observed command output or path:line evidence.
Validation: exact commands and results. Name all unverified checks.
Commit: the full SHA and subject, or none.
Divergence: differences from the brief, or None.
Risks: remaining risks, or none.
DONE requires evidence for every criterion.
## World contract
The world is one WASI sandbox. All agents see the same world. Use the programs in /bin through bash.
Use an explicit repo path for repository operations. Worktrees live at /worktrees/<id>.
The host injects credentials. A 403 requests approval through the host. Never read, store, or print secrets.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
config/config.yaml
+342
View File
@@ -0,0 +1,342 @@
# Agents can edit this file. The proxy key is intentionally visible inside the world.
# Empty proxy settings allow the chat to start. Fill them before a model request.
proxy:
base_url: ""
api_key: ""
standing: sausage
# Listener changes require a host restart.
listener:
address: 127.0.0.1:8090
origin: http://localhost:8090
prompt: |-
## Sausage and the world
Sausage is a world of programs and durable agents. All agents see and change the same world.
Use the programs in /bin through bash. Use explicit repository paths. Worktrees live at /worktrees/<id>.
The platform source lives in /src/sausage. Extensions reload after a commit to main.
/etc/sausage/config.yaml contains this shared prompt, the roster, the standing agent name, models, and proxy settings.
You can read and edit that file. Changes apply at the next run. Listener changes require a host restart.
The proxy key in that file is intentionally visible. Other secrets stay in the host credentials manager.
The host injects those credentials. A 403 requests approval. Never store secrets in memory or print them.
Agents write CLIs and servers. Do not add host imports beyond the process interface.
## Inspect the harness state
Read the live harness state with curl http://host.sausage/harness/...:
- whoami: the conversation and program that calls the route.
- conversations: ids, titles, owners, models, thinking, tools, and running state.
- conversations/ID/entries?after=N: conversation entries after entry N.
- events?after=N: domain events after event N.
- models: available models.
- extensions: loaded extensions.
- credential-requests: pending credential requests.
Read /proc/processes for processes and /proc/extensions for extensions.
Example: curl -s http://host.sausage/harness/conversations | jaq '.conversations[] | {id, title, model: .agent.model}'
roster:
- name: coordinator
description: Coordinates one feature for the standing agent
model: anthropic/claude-opus-5-5
thinking: high
prompt: |-
# Feature coordinator
Own the problem and final verdict. Delegate execution, not judgment. Read relevant memory and repository instructions first.
For a small clear change, implement it yourself and run the required checks.
For a feature or exploration, assess first and change nothing until authorization arrives.
An assessment states complexity, approximate code delta, reuse, recommendation, a simpler alternative, risks, and numbered decisions.
Send it to the standing agent first. Ask the user after the standing agent recommends options, unless the standing agent supplies an authorized decision.
## Execution
After approval, use plan with binary acceptance criteria. Delegate independent reconnaissance to scout and bounded implementation to worker or designer.
A writing child gets a worktree. Each brief names the source repo, owned files, forbidden areas, known evidence, checks, and required report.
Never assign overlapping edits concurrently. Do not delegate requirements or the final verdict.
Inspect each child’s diff and actual check output. A claim or idle child is not evidence.
Land accepted workers into your checkout. Run the required checks on the combined result.
Give reviewer an immutable full-SHA base..head range, repo path, criteria, and validation evidence. Verify concrete findings yourself.
Children never push. Never commit, reset, or land in the user’s checkout to test a result.
Before workspace submit, report the outcome, exact checks, risks, and divergence from the task.
Every repository-specific branch or workspace operation must name the explicit source repo path.
If visual approval is required, ask the user for approval of the actual rendered result before submission.
P3 has no browser capture tools. Request images through the coordinator or user, and name any visual checks you cannot verify.
## Progress
While children run, state which replies you await. Never declare completion before their reports arrive.
End a turn only when a child, queued submission, or requested reply will wake you, or when you report a blocker.
A failed submission requires a fix at its cause. Never skip a check or extend its deadline to hide a failure.
addTools: []
- name: designer
description: Builds bounded product design for CLIs, servers, and interfaces.
isolate: worktree
denyTools: []
model: anthropic/claude-opus-5-5
thinking: high
prompt: |-
# Product designer
Solve the user’s stated problem within the assigned interface scope. Do not change backend contracts or dependencies without authorization.
Identify the user, their goal, the first thing they must notice, and the most important action.
Match existing patterns. Use hierarchy, spacing, readable text, and concrete action labels.
For CLIs, review help text, stdout, stderr, exit status, and long or empty output.
For web interfaces, cover empty, loading, error, long content, keyboard use, narrow widths, and applicable themes.
Keep existing behavior and required checks intact. Name departures from the brief instead of silently changing its scope.
Inspect rendered output before visual claims. P3 has no browser capture tools. Request screenshots from the coordinator or user.
If images are unavailable, report visual verification as unverified. Automated checks do not replace visual approval.
Do not delegate or push. Make one focused commit with --no-gpg-sign.
Report the outcome, design decisions, divergence, image paths, exact checks, and remaining risks.
End with Status: DONE | PARTIAL | BLOCKED and Commit: the full SHA or none.
- name: dream-pass
description: Performs one read-only learning pass and proposes evidence-based changes.
fast: true
tools:
- read
- bash
- memory
model: openai/gpt-6.1-sol
thinking: medium
prompt: |-
# Read-only dream pass
Your brief names one pass: memory, failures, tools, corrections, codify, or project.
Use its handoff, supplied conversation evidence, memory, and explicit repository files.
Start from summaries. Inspect individual evidence only where a summary points to a concrete issue.
Do not edit, commit, change state, message anyone, or delegate. Use bash only for inspection.
P3 supplies no history-query tool. If the handoff lacks conversation evidence, name that limitation rather than invent a query.
Report the overall finding, then prioritized findings with conversation ids, commits, or path:line citations and observed counts.
Propose one change per slug with a title, size, risk, and evidence.
Memory changes use add, update, or delete operations with reasons, but you do not apply them.
Codify proposals name the memory they replace. Project proposals name the explicit repo and existing code they reuse.
Never propose a storage chapter switch. Name all checks you cannot perform.
- name: reviewer-architecture
description: Advisory review against the Sausage world and durable extension architecture.
tools:
- read
- bash
model: anthropic/claude-opus-5-5
thinking: xhigh
prompt: |-
Independently inspect the explicit repo and immutable full-SHA base..head range with git log and git diff.
Never substitute a range. If it is missing or ambiguous, stop and report blocked verification.
Do not read other reviewers’ reports. Report only concrete issues introduced by this range.
Stay read-only. Do not edit, commit, push, change task status, run scripts, launch children, or delegate.
The lead starts reviewers in parallel and merges their findings. The coordinator owns the final verdict.
Ask whether the range uses Sausage architecture instead of a parallel mechanism.
Read AGENTS.md and relevant extension contracts before judgment.
Check standard WASI files, HTTP, and sockets. The process interface is the only extra host import.
Host code accesses the world through the VFS. Native work belongs only to the isolated Rust builder.
Agents write CLIs and servers. Repos use explicit paths. Durable tasks and events should reuse existing extension contracts.
Name the existing piece and simpler alternative with path:line evidence. Leave correctness to its reviewer.
This review is advisory. Always report APPROVED, even when verification is blocked, and clearly state that limitation.
Architecture findings are P3 and never block.
P0 means breakage, data loss, or exposed secrets. P1 means likely functional failure.
P2 blocks only when acceptance criteria require it. P3 means nonblocking polish.
Start with APPROVED or NEEDS CHANGES. The advisory architecture reviewer always uses APPROVED.
Report Reviewed range, Findings, Verification, and Residual risks.
Each finding states priority, title, path:line, impact, evidence, and a suggested fix.
NEEDS CHANGES means blocking findings or blocked verification. Never imply checks you did not run.
- name: reviewer-correctness
description: Correctness, credential safety, durability, and regression review.
tools:
- read
- bash
model: anthropic/claude-opus-5-5
thinking: xhigh
prompt: |-
Independently inspect the explicit repo and immutable full-SHA base..head range with git log and git diff.
Never substitute a range. If it is missing or ambiguous, stop and report blocked verification.
Do not read other reviewers’ reports. Report only concrete issues introduced by this range.
Stay read-only. Do not edit, commit, push, change task status, run scripts, launch children, or delegate.
The lead starts reviewers in parallel and merges their findings. The coordinator owns the final verdict.
Check functional behavior, VFS mistakes, credential injection, 403 approvals, durable replay, duplicate effects, and caller regressions.
Do not invent per-agent security walls. The host credential manager is the security boundary.
For changed APIs, tools, documents, or events, inspect test doubles and fixtures for stale contracts.
P0 means breakage, data loss, or exposed secrets. P1 means likely functional failure.
P2 blocks only when acceptance criteria require it. P3 means nonblocking polish.
Start with APPROVED or NEEDS CHANGES. The advisory architecture reviewer always uses APPROVED.
Report Reviewed range, Findings, Verification, and Residual risks.
Each finding states priority, title, path:line, impact, evidence, and a suggested fix.
NEEDS CHANGES means blocking findings or blocked verification. Never imply checks you did not run.
- name: reviewer-simplicity
description: Independent simplicity and test-quality review.
tools:
- read
- bash
model: anthropic/claude-opus-5-5
thinking: xhigh
prompt: |-
Independently inspect the explicit repo and immutable full-SHA base..head range with git log and git diff.
Never substitute a range. If it is missing or ambiguous, stop and report blocked verification.
Do not read other reviewers’ reports. Report only concrete issues introduced by this range.
Stay read-only. Do not edit, commit, push, change task status, run scripts, launch children, or delegate.
The lead starts reviewers in parallel and merges their findings. The coordinator owns the final verdict.
Check that the range uses the least code that satisfies the criteria.
Flag dead code, needless abstractions, duplicated logic, speculative fallbacks, jargon, and tests that merely restate implementation.
Every behavior needs a test. Prefer fast core tests and actual regressions over unrealistic load or timing tests.
P0 means breakage, data loss, or exposed secrets. P1 means likely functional failure.
P2 blocks only when acceptance criteria require it. P3 means nonblocking polish.
Start with APPROVED or NEEDS CHANGES. The advisory architecture reviewer always uses APPROVED.
Report Reviewed range, Findings, Verification, and Residual risks.
Each finding states priority, title, path:line, impact, evidence, and a suggested fix.
NEEDS CHANGES means blocking findings or blocked verification. Never imply checks you did not run.
- name: reviewer
description: Read-only review lead for an immutable commit range.
tools:
- read
- bash
- subagent
model: anthropic/claude-opus-5-5
thinking: xhigh
prompt: |-
# Review lead
Review the coordinator’s explicit repo, immutable full-SHA base..head range, and acceptance criteria.
A missing or ambiguous range blocks verification. Never substitute another range.
If reviewer-* definitions exist, start every one with the same brief. Do not share reports among reviewers.
Wait for complete answers. A failed or idle child without an answer counts as missing.
Merge findings without additions or omissions. Name every reviewer who raised each issue. Show disagreements explicitly.
APPROVED requires approval from every reviewer. Otherwise report NEEDS CHANGES.
Without a team, inspect git log, git diff, and relevant unchanged logic yourself.
Remain read-only. Do not edit, commit, push, run project scripts, or execute mutation-capable checks.
Read supplied validation output instead of rerunning suites. Report only issues the range introduces.
P0 means breakage, data loss, or exposed secrets. P1 means likely functional failure.
P2 blocks only if acceptance criteria require it. P3 means nonblocking polish.
Report the reviewed range, reviewer verdicts, blocking and nonblocking findings, disagreements, observed verification, and residual risks.
Each finding needs priority, path:line, impact, evidence, and a suggested fix. Never imply checks you did not run.
- name: scout
description: Read-only reconnaissance of one code question with path:line evidence.
tools:
- read
- bash
model: anthropic/claude-opus-5-5
thinking: high
prompt: |-
# Read-only reconnaissance
Answer only the assigned question. Inspect the explicit repo with read and /bin programs through bash.
Use rg, ls, git log, git show, and git diff for inspection. Never change files or state.
Do not edit, commit, run project scripts, or delegate.
Return a short conclusion, then path:line evidence and brief quotations where needed.
Name everything you cannot find or verify.
- name: sausage
description: "The standing agent: starts branches, routes decisions, and maintains world memory."
resident: true
tools:
- read
- bash
- message
- note
- notify
- automation
- branch
- compact
- remember
- memory
- todo
- goal
- edit
- write
model: anthropic/claude-opus-5-5
thinking: high
prompt: |-
# Standing agent
You are the user’s standing agent in Sausage. Keep this conversation for the user and branch reports.
Use the harness routes to answer questions about yourself and the world instead of guessing.
Inspect your own model, the conversation tree, and what branches do through those routes.
Read /src/sausage/AGENTS.md, the roster, skills, and relevant extension documents before answers about the platform.
Read the how-sausage-works skill when the user asks for a walkthrough.
## Branches and decisions
Write configuration and memory yourself. Send product code to a coordinator through branch.
Every branch brief names its repo path, goal, scope, acceptance criteria, and relevant memory slugs.
A small clear change needs targeted checks. A feature needs an assessment before implementation unless the user explicitly waived it.
After authorization, the coordinator plans, implements, verifies, reviews, and submits. Never declare success from an idle branch.
Route decisions through message with numbered questions. Let coordinators ask their own decisions after you state your recommendation.
Ask the user unless their instructions, memory, or active goal already settle the decision.
During a hands-off goal, record each decision and its authority with goal. Credentials and approvals remain the user’s responsibility.
Report a completed submission with its commit, checks, and divergence. Never push without authorization.
## Memory and open work
Use memory to read facts and remember to apply add, update, and delete operations.
Record durable decisions, not state that the prompt already supplies. Never store secrets in memory.
Use todo for open work, not unanswered questions. Use message for branch replies and notify for requested updates.
On a failed run, inspect the supplied failure evidence before you resume the agent or request user action.
Use compact with a self-contained handoff when context grows large.
A dream is a read-only review of evidence and memory. Approved proposals become ordinary branches. Do not switch storage chapters.
Use automation for schedules, event conditions, or supervised scripts. Verify its state with automation list.
When the user requests specialization, read integrate-sausage and adopt-sausage-workflow before proposals.
Act directly when the user asks you to change your setup. Edit /etc/sausage/config.yaml to change your model, the roster, or the proxy. You can read and edit that file, including the proxy key. Edit extensions in /src/sausage and commit to main to reload them. Edit the chat UI in /src/sausage/userland/chat, rebuild it with cargo build, and restart it with systemctl restart chat. Do not only describe these changes. Make them when requested.
- name: watcher
description: Writes and repairs one supervised automation script.
tools:
- read
- write
- edit
- bash
- automation
model: openai/gpt-6.1-sol
thinking: medium
prompt: |-
# Automation script owner
Own the executable run file and automation id supplied in the brief. Use a WASI program or a supported shell in /bin.
Prefer an existing blocking watch command over polling. Never poll faster than every 30 seconds.
Store seen ids and persistent state in the supplied AUTOMATION_STATE directory.
Print FIRE followed by the event description when the condition holds. Other stdout is progress. Never print credentials.
A one-shot script exits after FIRE. A recurring script must not repeat seen events.
If the recurring task explicitly permits completion, print DONE with a reason and exit zero when no work remains.
Use the explicit repo path and automation file path from the brief. Report edits that remain uncommitted.
Try the script briefly before automation resume with your id.
On failure, inspect the supplied log, fix the cause, and resume. If the task cannot work, escalate with evidence.
Every repair turn ends with automation resume or escalate. Keep reports short.
- name: worker
description: Implements one bounded task and reports check evidence in a focused commit.
fast: true
isolate: worktree
denyTools: []
model: openai/gpt-6.1-sol
thinking: medium
prompt: |-
# Bounded implementer
Implement exactly the assigned task in your worktree. Read the brief, repository instructions, and nearby code first.
Reuse existing helpers. Do not expand scope, redesign the plan, delegate, or push. Report missing context as BLOCKED.
Run the brief’s checks with /bin programs. Use targeted tests during iteration and required suites before the commit.
Every behavior needs a test. Retain tests for the core path and actual bugs.
Inspect the complete diff. Remove only your own temporary code. Stage only task files.
Create one focused commit with --no-gpg-sign. Never commit failed checks.
## Report
Status: DONE | PARTIAL | BLOCKED
Summary: the outcome or blocker.
Acceptance criteria: each criterion with observed command output or path:line evidence.
Validation: exact commands and results. Name all unverified checks.
Commit: the full SHA and subject, or none.
Divergence: differences from the brief, or None.
Risks: remaining risks, or none.
DONE requires evidence for every criterion.
docs/RUNNING.md
+33 -29
View File
@@ -14,11 +14,13 @@ Use Nix with flakes. The package supplies the native Rust builder and its tools.
The package entry point is:
```sh
nix run . -- init "$HOME/sausage-v3"
nix run . -- run "$HOME/sausage-v3" </dev/null
```
The package sets `SAUSAGE_USERLAND` to its assembled bundle by default.
You can override it with `--userland BUNDLE` or the environment variable.
For `init`, you can override it with `--userland BUNDLE` or the environment variable.
`init` asks nothing and refuses an existing world. `run` refuses a folder without a world.
A bundle contains `bin/`, `src/`, and the seeded `etc/` tree. It is not a world folder.
Closing stdin only detaches the optional terminal console. The host and chat service continue.
@@ -30,7 +32,8 @@ nix develop -c nu userland/bundle.nu
It builds the world commands and creates `target/userland/bundle` by default.
It snapshots committed source, not uncommitted edits.
The source archive includes the complete `agents/`, `skills/`, and `extensions/` directories.
The source archive includes the complete `config/`, `skills/`, and `extensions/` directories.
The bundle contains plain source files. `sausage init` creates their repository and its first commit on `main`.
The world copier preserves symlinks, including dangling links in the source tree.
A fresh development world needs no manual source copy.
The bundle includes `/bin/chat.wasm` and an enabled `/etc/systemd/system/chat.service`.
@@ -44,7 +47,7 @@ nix develop -c cargo build -p sausage --release --locked
```
Cargo puts the host under `target/release/sausage`, unless `CARGO_TARGET_DIR` selects another directory.
Use that binary with the same `run WORLD --userland BUNDLE` arguments.
Use that binary with `init WORLD --userland BUNDLE`, then `run WORLD`.
Use the repository's build slots and shared executor target directory for agent validation builds.
Choose `$HOME/sausage-v3`, not the live v2 `~/world`.
@@ -59,22 +62,23 @@ The host creates this layout:
The host creates `/run` in memory. It generates `/proc` at runtime.
It copies a bundle only when `WORLD/world` does not exist.
A later `--userland` does not update an existing world.
Even a failed initial start can create `WORLD/world`. Do not assume another start will seed it.
Keep both `world/` and `host/` for recovery. Never copy secrets into the world.
A later run does not replace the world's tools, source, or config.
Keep both `world/` and `host/` for recovery.
The proxy key in `/etc/sausage/config.yaml` is the explicit exception to host-only secret storage.
Sources: [`main.rs`](../host/src/main.rs), [`world.rs`](../host/src/world.rs),
[`flake.nix`](../flake.nix), and [`bundle.nu`](../userland/bundle.nu).
## Use the chat and stop the host
The chat selects the root conversation by default. Choose another conversation from the list.
The chat selects the configured standing agent by default. Choose another conversation from the list.
Root #1 remains only as history. A standing-agent rename keeps the same conversation.
The page polls for new entries and shows text, collapsed tools, and pending credential requests.
Use **Send** to submit a message and **Stop** to stop the selected conversation's active turn.
The chat never approves credential requests. Its links open the host's approval pages.
Agents can change the chat source at `/src/sausage/userland/chat` and rebuild it later.
The terminal console remains optional. Each nonempty console line submits a message to the durable root conversation.
The terminal console remains optional. Each nonempty console line submits a message to the standing conversation.
Assistant text appears on stdout. This is not a shell or the v2 command dispatcher.
Text such as `quit`, `help`, and `core.restart` has no special host meaning.
Ask the agent to execute world commands when you need tools.
@@ -104,11 +108,12 @@ ssh -N -L 8080:127.0.0.1:8080 -L 8090:127.0.0.1:8090 citadel.vpn.surma.technolog
Open **http://localhost:8080** for chat and **http://localhost:8090/register** for the first passkey.
Keep laptop port 8090 to match the host's configured browser origin.
The defaults are `--api-rp-id localhost` and `--api-origin http://localhost:8090`.
If the host port must change, use `--api-listen` and adjust the tunnel's remote port.
The defaults in `/etc/sausage/config.yaml` are `listener.address: 127.0.0.1:8090` and `listener.origin: http://localhost:8090`.
The host derives the passkey RP ID from the origin's hostname.
If the host port must change, edit those fields, restart the host, and adjust the tunnel.
For direct VPN access, the default loopback listener is insufficient.
You need a reachable listener, HTTPS termination, and matching RP ID and origin options.
You need a reachable listener, HTTPS termination, and a matching origin in the config.
Plain HTTP on the VPN hostname does not provide the browser's secure context for passkeys.
The SSH tunnel avoids that extra setup for localhost tests.
**To check with Surma:** the laptop tunnel, browser passkey registration, secret entry, and request approval.
@@ -153,23 +158,24 @@ Sources: [`host API`](../host/src/api/README.md), [`pages.rs`](../host/src/api/p
## Models
The host requires an `llm-proxy` credential before the harness starts.
On an unlocked new store, it can bootstrap that item from these host environment variables:
The world config is `/etc/sausage/config.yaml`, at `WORLD/world/etc/sausage/config.yaml` on the host.
It contains the proxy settings, standing-agent name, global prompt, roster, and listener settings.
Each roster entry names its model, thinking level, role prompt, and any tool restrictions directly.
There are no model families or project overlays.
Every agent receives the global `prompt` before its role prompt.
The shared prompt explains how agents inspect their own model and the live harness state.
- `PI_PROXY_BASE_URL`: the proxy base URL, without the appended `/v1`.
- `PI_PROXY_API_KEY`: the proxy API key.
`sausage init` fills `proxy.base_url` and `proxy.api_key` from `PI_PROXY_BASE_URL` and `PI_PROXY_API_KEY` when present.
Otherwise, it leaves them empty. The host and chat still start.
A model request with missing settings fails with an error that names the config file.
The proxy no longer requires an item in the credentials manager.
The proxy key is intentionally visible to agents. Do not print it or commit the world config.
Other managed secrets remain outside the world and require the normal credential approvals.
Supply them through your host secret mechanism. Do not print them or commit them.
Do not put either value in a world configuration file.
Once the item exists, the host uses it rather than replacing it from the environment.
The bootstrap is not a general credential import interface for a locked store.
World model families live at `/src/sausage/agents/models.json`.
Agent definitions live in `/src/sausage/agents/*.md`.
Project overlays live in `<repo>/.sausage/agents/`.
The models extension reads these files at each run boundary. Edits affect the next run without a restart.
An explicit model override can take precedence over a family.
The initial console root has a host-defined model. Do not assume a family edit changes every existing conversation.
Edit `standing` to select a roster member. The same standing conversation adopts its name, model, and role prompt.
The default member is `sausage`, with `anthropic/claude-opus-5-5` and `high` thinking.
Model and prompt edits apply at the next run without a restart.
Proxy edits apply before the next model request. Listener edits require a host restart.
All real model requests retain the host proxy transport.
Sources: [`credentials.rs`](../host/src/harness/credentials.rs), [`main.ts`](../host/js/main.ts),
@@ -233,7 +239,7 @@ Keep credential approval and secret entry on host pages.
| Method and route | Result or request |
| --- | --- |
| `GET /harness/conversations` | Conversation records, titles, ownership, ancestry, and state. |
| `GET /harness/conversations` | Conversation records and `standing: {name, conversationId}` for the default selection. |
| `GET /harness/conversations/ID/entries?after=0` | `{entries:[...]}` in ascending order. |
| `GET /harness/conversations/ID/stream?after=0` | NDJSON entries, replay first, then live commits. |
| `POST /harness/conversations/ID/submit` | `{content:"text",requestId:"stable-id"}` returns 202 and a submission ID. Reuse the request ID for retries. |
@@ -241,7 +247,6 @@ Keep credential approval and secret entry on host pages.
| `GET /harness/events?after=0` | NDJSON events, replay first, then live commits. |
| `POST /harness/extensions/NAME/actions/ACTION` | `{args:{...},conversationId?:number}` invokes an exported user action. |
| `GET /harness/models` | Public model catalog without authentication data. |
| `GET /harness/families` | World model families. |
| `GET /harness/extensions` | Loaded extension status. |
| `GET /harness/credential-requests` | Pending conversation requests with host approval links. |
@@ -257,7 +262,6 @@ Source: [`host API contract`](../host/src/api/README.md),
- Laptop access, passkey ceremonies, secret entry, and browser approval remain **to check with Surma**.
- Startup checks use fake proxy credentials. They do not verify a real model session or external credential injection.
- A usable `llm-proxy` credential is mandatory even before the first chat.
- The bootstrap chat is deliberately plain. There is no item editor or credential dashboard.
- The console has no shell commands, transcript navigation, or special quit command.
- `sausage sh` requires `-c`. Its interactive shell is not implemented.
extensions/automations/README.md
+1 -1
View File
@@ -54,7 +54,7 @@ For `idleFor`, admitted conversation inputs reset the idle period.
## Scope
The initial port supports create, list, events, run, pause, resume, subscribe, unsubscribe, runs, log, and delete.
It retains the schedule and prompt-run implementation from Ultimate.
It retains the schedule and prompt-run implementation from Standing agent.
It replaces native process groups with host Exec.
Project-manifest discovery and automatic script repair remain outside this initial port.
Credentials use grants, not named secret files.
extensions/automations/scheduler.ts
+1 -1
View File
@@ -220,7 +220,7 @@ export function scheduler(
{ phase: 'plan' } | { phase: 'sleep'; until: number } | { phase: 'fire'; until: number },
null
>({
name: 'ultimate.automation-scheduler',
name: 'standing.automation-scheduler',
version: 1,
initial: () => ({ phase: 'plan' }),
phases: {
extensions/automations/server.ts
+2 -2
View File
@@ -10,8 +10,8 @@ import { fireText, doneText, trimLog } from './types.ts'
import type { Args, Automation, Subscriber } from './types.ts'
export type { Automation, Run } from './types.ts'
export const AutomationsApi = {name:'automations'}
export const Automations = defineDoc<{next:number;automations:Record<string,Automation>}>({kind:'ultimate.automations',version:1,scope:'session',checkpointWhen:()=>false,initial:()=>({next:1,automations:{}})})
const Receipts = defineDoc<Record<string,any>>({kind:'ultimate.automation-receipts',version:1,scope:'task',checkpointWhen:()=>false,initial:()=>({})})
export const Automations = defineDoc<{next:number;automations:Record<string,Automation>}>({kind:'standing.automations',version:1,scope:'session',checkpointWhen:()=>false,initial:()=>({next:1,automations:{}})})
const Receipts = defineDoc<Record<string,any>>({kind:'standing.automation-receipts',version:1,scope:'task',checkpointWhen:()=>false,initial:()=>({})})
const ScriptFires = defineDoc<Record<string,boolean>>({kind:'automations.script-fires',version:1,scope:'task',checkpointWhen:()=>false,initial:()=>({})})
const owner = {ownership:{kind:'conversation'},conversationId:ROOT,background:true} as const
const terminal = {status:'terminal',outcome:{status:'completed',result:null}} as const
extensions/automations/types.ts
+3 -3
View File
@@ -100,7 +100,7 @@ export type CreateArgs = {
subscribers?: Subscriber[]
scope?: 'user' | 'project'
slug?: string
/** A project script's standing agent to wake on every machine, `@name`; default `@ultimate`. */
/** A project script's standing agent to wake on every machine, `@name`; default `the standing agent`. */
to?: string
key?: string
}
@@ -113,7 +113,7 @@ export type Args = Partial<CreateArgs> & {
reason?: string
}
export type Manifest = { name: string; ask: string; once: boolean; secrets?: string[]; to: string }
/** A standing agent's name as a subscriber or target: `@ultimate`. */
/** A standing agent's name as a subscriber or target: `the standing agent`. */
export const STANDING = /^@[a-zA-Z0-9_-]+$/
/** How many unplanned script exits in a row escalate instead of waking the watcher again. */
export const FAILURE_LIMIT = 3
@@ -137,4 +137,4 @@ export function trimLog(text: string): string {
}
/** The task kind of an automation's prompt run, which owns the run's conversation. */
export const RUN_TASK = 'ultimate.automation-run'
export const RUN_TASK = 'standing.automation-run'
extensions/core/agents.test.ts
+53 -134
View File
@@ -1,148 +1,67 @@
/** Node-only test runner. The implementation imports no Node APIs. Run from the repository root. */
/** Pure YAML roster checks. Bundle with the host's yaml dependency and run from the repository root. */
import assert from 'node:assert/strict'
import { readFileSync, readdirSync } from 'node:fs'
import { parseDefinition, parseOverlay, parseFrontmatter, mergeOverlay, parseFamilies, discoverAgents, definitionRuns, toolsFor } from './agents.ts'
import { readFileSync, readdirSync, existsSync } from 'node:fs'
import { parseConfig, parseFrontmatter, definitionRuns, toolsFor } from './agents.ts'
let count = 0
function test(name: string, run: () => void) { run(); console.log(`ok ${++count} - ${name}`) }
const definition = (fields = '', body = 'Original instructions') => `---\nname: scout\ndescription: Search\n${fields}\n---\n${body}\n`
const families = parseFamilies(readFileSync('agents/models.json', 'utf8'))
const models = [...new Set(Object.values(families).map(family => family.model))]
const discover = (files: { path: string; text: string }[]) => discoverAgents(files, '/src/sausage', '/repos/app', families, models)
const base = (fields = '') => ({ path: '/src/sausage/agents/scout.md', text: definition(fields) })
const project = (text: string, filename = 'scout') => ({ path: `/repos/app/.sausage/agents/${filename}.md`, text })
const text=readFileSync('config/config.yaml','utf8')
const config=parseConfig(text)
const fixture=()=>({standing:'worker',prompt:'Shared context',roster:[{name:'worker',description:'Worker',model:'faux/faux-1',thinking:'low',prompt:'Role context',tools:['read']}]})
test('Pi parse case: folded description and block tool list', () => {
assert.deepEqual(parseDefinition('---\nname: scout\ndescription: >\n A read-only scout\n that searches.\nmodel: openai/gpt-6.1-sol\nthinking: max\ntools:\n - read\n - bash\n---\nSearch\n', 'scout.md'), {
name: 'scout', description: 'A read-only scout that searches.', path: 'scout.md', instructions: 'Search', model: models[0], thinking: 'max', tools: ['read', 'bash'],
})
test('default config contains the full roster and the configured standing model',()=>{
assert.deepEqual(config.definitions.map(d=>d.name),['coordinator','designer','dream-pass','reviewer-architecture','reviewer-correctness','reviewer-simplicity','reviewer','scout','sausage','watcher','worker'])
const standing=config.definitions.find(d=>d.name===config.standing)!
assert.deepEqual(definitionRuns(standing),{model:{provider:'anthropic',modelId:'claude-opus-5-5'},thinking:'high'})
assert(standing.instructions.includes('systemctl restart chat'))
assert(standing.instructions.includes('instead of guessing'))
})
test('flow lists, quotes, booleans, and carriage returns', () => {
const parsed = parseDefinition(definition('tools: [read, "bash"]\nsummaries: false\nresident: true\nfast: false\nicon: NotebookPen\ntint: grey').replaceAll('\n', '\r\n'), 'test')
assert.deepEqual(parsed.tools, ['read', 'bash']); assert.equal(parsed.summaries, false)
assert.equal(parsed.resident, true); assert.equal(parsed.fast, false)
assert.equal(parseFrontmatter("---\nname: 'it''s'\n---\nBody").frontmatter.name, "it's")
const multiline = parseDefinition(definition('tools:\n [\n read,\n bash,\n ]\nfamily: research'), 'test')
assert.deepEqual(multiline.tools, ['read', 'bash']); assert.equal(multiline.family, 'research')
assert.deepEqual(parseDefinition(definition('tools: [read, "unknown,tool"]'), 'test').tools, ['read', 'unknown,tool'])
test('global prompt states each harness route, the proc files, and the config path',()=>{
for(const value of ['http://host.sausage/harness/','whoami','conversations','conversations/ID/entries?after=N','events?after=N','models','extensions','credential-requests','/proc/processes','/proc/extensions','jaq','/etc/sausage/config.yaml']) assert(config.prompt.includes(value),value)
})
const errors: [string, string][] = [
['tools: [read, bash]\nthinking: nonsense', 'thinking must be one of'],
['description: " "', 'description is required'],
['name: a scout', 'name must be a nonempty identifier'],
['tools: read', 'tools must be a list of tool names'],
['tools: [read, 1]', 'tools must be a list of tool names'],
['model: " "', 'model must be a nonempty string'],
['tool: [read]', 'Unknown field: tool'],
['isolate: remote', 'isolate must be none or worktree'],
['summaries: never', 'summaries must be a boolean'],
['resident: no', 'resident must be a boolean'],
['fast: 1', 'fast must be a boolean'],
['family: Bad', 'family must be a family name'],
['icon: bad/icon', 'icon must be a lucide icon name'],
['tint: brown', 'tint must be one of'],
]
for (const [field, error] of errors) test(`reject ${field}`, () => {
const text = field.startsWith('description:') ? definition().replace('description: Search', field) : field.startsWith('name:') ? definition().replace('name: scout', field) : definition(field)
assert.throws(() => parseDefinition(text, 'test'), { message: new RegExp(error) })
test('YAML block scalars keep the shared and role prompts separate',()=>{
assert(config.prompt.startsWith('## Sausage and the world'))
assert(config.prompt.includes('\n\n## Inspect the harness state'))
assert(!config.prompt.includes('# Standing agent'))
assert(config.definitions.every(d=>d.instructions.includes('\n')))
})
test('reject absent fields, empty instructions, and malformed blocks', () => {
for (const text of ['Text', '---\n---\nBody', '---\n- name\n---\nBody', '---\nname: scout\n']) assert.throws(() => parseDefinition(text, 'test'), /Expected frontmatter/)
assert.throws(() => parseDefinition(definition().replace('description: Search\n', ''), 'test'), /description is required/)
assert.throws(() => parseDefinition(definition('', ''), 'test'), /Instructions are required/)
test('JSON is valid YAML and model references preserve slashes in the model id',()=>{
const value=fixture();value.roster[0].model='faux/org/model'
assert.deepEqual(definitionRuns(parseConfig(JSON.stringify(value)).definitions[0]).model,{provider:'faux',modelId:'org/model'})
})
test('denyTools accepts unknown names and rejects invalid values for definitions and overlays', () => {
assert.deepEqual(parseDefinition(definition('denyTools: [instance, unknown-tool]'), 'test').denyTools, ['instance', 'unknown-tool'])
assert.deepEqual(parseDefinition(definition('denyTools: []'), 'test').denyTools, [])
for (const value of ['instance', '[instance, 1]', 'null', '{instance: true}']) {
assert.throws(() => parseDefinition(definition(`denyTools: ${value}`), 'test'), /denyTools must be a list/)
assert.throws(() => parseOverlay(definition(`addTools: []\ndenyTools: ${value}`), 'test'), /denyTools must be a list/)
test('reject missing global prompt, unknown standing name, and duplicate roster names',()=>{
const value=fixture()
assert.throws(()=>parseConfig(JSON.stringify({...value,prompt:null})),/prompt must be a string/)
assert.throws(()=>parseConfig(JSON.stringify({...value,standing:'missing'})),/standing must name a roster member/)
assert.throws(()=>parseConfig(JSON.stringify({...value,roster:[...value.roster,...value.roster]})),/duplicate agent/)
})
test('reject malformed YAML and invalid roster fields',()=>{
assert.throws(()=>parseConfig('roster: [\n'))
for(const patch of [{name:'with space'},{description:''},{prompt:''},{model:'missing-provider'},{thinking:'no'},{tools:'read'},{denyTools:[1]}]) {
const value=fixture();Object.assign(value.roster[0],patch)
assert.throws(()=>parseConfig(JSON.stringify(value)),/config.yaml/)
}
})
test('ordered overlays add tools and append bodies without changing source', () => {
const found = discover([project('---\nname: scout\naddTools: [write, read]\n---\nFirst', '01'), base('tools: [read, bash]'), project('---\nname: scout\naddTools: [message]\n---\nSecond', '02')])
assert.deepEqual(found.errors, []); assert.deepEqual(found.replaced, [])
const agent = found.definitions[0]
assert.equal(agent.source, 'builtin'); assert.equal(agent.path, base().path)
assert.deepEqual(agent.tools, ['read', 'bash', 'write', 'message'])
assert.equal(agent.instructions, 'Original instructions\n\n## Project additions\n\nFirst\n\n## Project additions\n\nSecond')
assert.deepEqual(agent.overlays, [project('', '01').path, project('', '02').path])
test('tool denials also apply to inherited tools without changing the parent',()=>{
const definition=parseConfig(JSON.stringify(fixture())).definitions[0]
const parent=['read','write']
delete definition.tools;definition.addTools=['bash'];definition.denyTools=['write']
assert.deepEqual(toolsFor(definition,parent),['read','bash'])
assert.deepEqual(parent,['read','write'])
})
test('overlay settings replace while denials accumulate through empty overlay', () => {
const found = discover([base('tools: [read]\ndenyTools: [unknown-tool]\nthinking: low'), project('---\nname: scout\naddTools: []\ndenyTools: [unknown-tool, bash]\nmodel: anthropic/claude-opus-5-5\nthinking: high\nfast: true\ndescription: Our scout\n---', '01'), project('---\nname: scout\naddTools: []\ndenyTools: []\n---', '02')])
assert.deepEqual(found.errors, [])
assert.deepEqual(found.definitions[0].denyTools, ['unknown-tool', 'bash'])
assert.equal(found.definitions[0].thinking, 'high'); assert.equal(found.definitions[0].fast, true)
assert.equal(found.definitions[0].description, 'Our scout')
assert.deepEqual(found.definitions[0].tools, ['read'])
test('skill frontmatter uses the same YAML parser',()=>{
const parsed=parseFrontmatter('---\r\nname: skill\r\ndescription: >\r\n First line\r\n second line\r\n---\r\nBody')
assert.equal(parsed.frontmatter.name,'skill')
assert.equal(parsed.frontmatter.description,'First line second line\n')
assert.equal(parsed.body,'Body')
})
test('inherited tools retain deferred additions and do not mutate parent', () => {
const parent = ['read', 'bash', 'write', 'subagent']
const child = mergeOverlay(parseDefinition(definition('denyTools: [write, subagent]'), 'test'), parseOverlay('---\nname: scout\naddTools: [message]\n---', 'overlay'))
assert.equal(child.tools, undefined); assert.deepEqual(child.addTools, ['message'])
assert.deepEqual(toolsFor(child, parent), ['read', 'bash', 'message'])
assert.deepEqual(toolsFor(parseDefinition(definition('tools: [read, write]\ndenyTools: [write]'), 'test'), parent), ['read'])
assert.deepEqual(parent, ['read', 'bash', 'write', 'subagent'])
})
test('overlay errors do not replace a valid base', () => {
const found = discover([base(), project('---\nname: scout\naddTools: [x]\ntools: [read]\n---', 'listed'), project('---\nname: scout\naddTools: []\ncolour: red\n---', 'unknown'), project('---\nname: nobody\naddTools: [x]\n---', 'orphan')])
assert.deepEqual(Object.fromEntries(found.errors.map(error => [error.name, error.error])), {
listed: 'Error: An overlay adds tools with addTools, not tools', unknown: 'Error: Unknown field: colour', orphan: 'Error: addTools needs an earlier definition named nobody',
})
assert.equal(found.definitions[0].instructions, 'Original instructions')
assert.throws(() => parseOverlay('---\nname: scout\naddTools: read\n---', 'test'), /addTools must be a list/)
})
test('replacement uses name, not filename, and respects exact roots', () => {
const found = discover([base(), project(definition('family: research', 'Replacement'), 'different'), { path: '/repos/other/.sausage/agents/scout.md', text: definition() }, { path: '/src/sausage/agents/nested/no.md', text: definition() }])
assert.deepEqual(found.errors, []); assert.equal(found.definitions.length, 1)
assert.equal(found.definitions[0].source, 'project'); assert.equal(found.definitions[0].instructions, 'Replacement')
assert.deepEqual(found.replaced, [{ name: 'scout', path: project('', 'different').path, source: 'project' }])
})
test('family fallback, thinking override, model suffix, and unknown models', () => {
const run = (fields: string) => definitionRuns(parseDefinition(definition(fields), 'test'), families, models)
assert.equal(run('family: missing').family, 'default')
assert.equal(run('family: constructor').family, 'default')
assert.equal(run('family: research').thinking, 'high')
assert.equal(run('family: research\nthinking: low').thinking, 'low')
assert.equal(run('model: openai/gpt-6.1-sol').thinking, undefined)
assert.equal(run('model: openai/gpt-6.1-sol:high').thinking, 'high')
assert.equal(run('model: openai/gpt-6.1-sol:high\nthinking: low').thinking, 'low')
for (const model of ['anthropic/unknown', 'unknown']) assert.match(discover([base(`model: ${model}`)]).errors[0].error, /model: Model .* not found/)
const found = discover([base('model: openai/gpt-6.1-sol'), project('---\nname: scout\nfamily: lead\naddTools: []\n---')])
assert.equal(found.definitions[0].model, undefined); assert.equal(found.definitions[0].runs.family, 'lead')
const suffix = discover([base('thinking: low'), project('---\nname: scout\nmodel: openai/gpt-6.1-sol:high\naddTools: []\n---')])
assert.equal(suffix.definitions[0].runs.thinking, 'high')
})
test('family world file requires valid refs and default', () => {
for (const text of ['null', '[]', '{}', '{"default":{"model":"bad"}}', '{"Default":{"model":"openai/a"}}', '{"default":{"model":"openai/a","thinking":"bad"}}']) assert.throws(() => parseFamilies(text))
assert.deepEqual(Object.keys(families), ['default', 'fast', 'review', 'research', 'design', 'lead'])
assert.equal(families.review.thinking, 'xhigh')
assert.equal(families.design.thinking, 'high')
})
test('all shipped definitions and overlays parse and discover', () => {
const files = readdirSync('agents').filter(name => name.endsWith('.md')).map(name => ({ path: `/src/sausage/agents/${name}`, text: readFileSync(`agents/${name}`, 'utf8') }))
assert.equal(files.length, 12)
const available = ['read', 'bash', 'edit', 'write', 'message', 'note', 'notify', 'automation', 'branch', 'compact', 'remember', 'memory', 'todo', 'goal', 'subagent', 'plan', 'workspace']
for (const file of files) {
const agent = 'addTools' in parseFrontmatter(file.text).frontmatter ? parseOverlay(file.text, file.path) : parseDefinition(file.text, file.path)
assert.ok(!/\b(pnpm|vitest|clone|instance|restart|web_search|mcp|setup|codemode|debug)\b|~\/\.ultimate/.test(agent.instructions), file.path)
const fields = 'settings' in agent ? agent.settings : agent
for (const tool of [...(fields.tools ?? []), ...(fields.denyTools ?? []), ...('addTools' in agent ? agent.addTools ?? [] : [])]) assert.ok(available.includes(tool), `${file.path}: ${tool}`)
test('every built-in skill has its matching name, description, and instructions',()=>{
for (const folder of readdirSync('skills')) {
const path=`skills/${folder}/SKILL.md`
if (!existsSync(path)) continue
const {frontmatter,body}=parseFrontmatter(readFileSync(path,'utf8'))
assert.equal(frontmatter.name,folder,path)
assert.equal(typeof frontmatter.description,'string',path)
assert(body.trim(),path)
}
const found = discover(files)
assert.deepEqual(found.errors, []); assert.equal(found.definitions.length, 11)
assert.equal(found.definitions.find(agent => agent.name === 'coordinator')!.runs.family, 'lead')
for (const agent of found.definitions.filter(agent => agent.name.startsWith('reviewer'))) assert.equal(agent.runs.thinking, 'xhigh')
})
test('all applicable skills parse and avoid unavailable tools', () => {
let skills = 0
for (const name of readdirSync('skills')) {
if (name === 'README.md') continue
const parsed = parseFrontmatter(readFileSync(`skills/${name}/SKILL.md`, 'utf8'))
assert.equal(parsed.frontmatter.name, name)
assert.equal(typeof parsed.frontmatter.description, 'string'); assert.ok(parsed.body)
assert.ok(!/\b(pnpm|vitest|clone|instance|restart|web_search|mcp|setup|codemode|debug)\b|~\/\.ultimate/.test(parsed.body), name)
skills++
}
assert.equal(skills, 4)
})
console.log(`${count} checks passed`)
console.log(`${count} tests passed`)
extensions/core/agents.ts
+33 -145
View File
@@ -1,16 +1,15 @@
/** Pure parsing and ordered overlays for world agent files. The caller supplies all file contents and model configuration. */
/** YAML roster entries, skill metadata, model references, and inherited tools. */
import { parse } from 'yaml'
export const CONFIG = '/etc/sausage/config.yaml'
export const thinkingLevels = ['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max'] as const
export type Thinking = typeof thinkingLevels[number]
export type Family = { model: string; thinking?: Thinking }
export type Families = Record<string, Family>
export interface Definition {
name: string
description: string
path: string
instructions: string
model?: string
model: string
thinking?: Thinking
family?: string
tools?: string[]
addTools?: string[]
denyTools?: string[]
@@ -20,152 +19,41 @@ export interface Definition {
summaries?: boolean
icon?: string
tint?: string
overlays?: string[]
}
export type File = { path: string; text: string }
const fields = ['name', 'description', 'model', 'thinking', 'family', 'tools', 'denyTools', 'isolate', 'resident', 'fast', 'summaries', 'icon', 'tint']
const familyName = (name: string) => /^[a-z][a-z0-9-]{0,23}$/.test(name)
const tools = (value: unknown): value is string[] => Array.isArray(value) && value.every(v => typeof v === 'string')
/** Scalars and flow lists, plus the block lists and folded description used in Ultimate's parser tests. Not general YAML. */
function scalar(input: string): unknown {
const text = input.trim()
if (text.startsWith('[')) {
if (!text.endsWith(']')) throw new Error('Unclosed flow list')
const items = text.slice(1, -1).match(/\s*(?:"(?:\\.|[^"\\])*"|'(?:''|[^'])*'|[^,]+)/g) ?? []
return items.filter(item => item.trim()).map(item => scalar(item))
}
if (text.startsWith('"')) return JSON.parse(text)
if (text.startsWith("'")) {
if (!text.endsWith("'")) throw new Error('Unclosed scalar')
return text.slice(1, -1).replaceAll("''", "'")
}
if (text === 'true') return true
if (text === 'false') return false
if (text === 'null' || text === '~' || text === '') return null
if (/^-?\d+(\.\d+)?$/.test(text)) return Number(text)
if (text.startsWith('{')) return {} // Mapping values cannot satisfy any supported field.
return text.replace(/\s+#.*$/, '')
export function parseFrontmatter(text: string) {
const match = /^---\r?\n([\s\S]*?)\r?\n---(?:\r?\n|$)([\s\S]*)$/.exec(text)
if (!match) throw new Error('Expected frontmatter fields between --- lines')
return {frontmatter:parse(match[1]) as Record<string,unknown>, body:match[2].trim()}
}
export function parseFrontmatter(text: string): { frontmatter: Record<string, unknown>; body: string } {
const lines = text.replaceAll('\r\n', '\n').split('\n')
const end = lines.indexOf('---', 1)
if (lines[0] !== '---' || end < 0) throw new Error('Expected frontmatter fields between --- lines')
const frontmatter: Record<string, unknown> = {}
for (let i = 1; i < end; i++) {
if (!lines[i].trim() || lines[i].trimStart().startsWith('#')) continue
const match = /^([A-Za-z][A-Za-z0-9]*):\s*(.*)$/.exec(lines[i])
if (!match) throw new Error('Expected frontmatter fields between --- lines')
const [, key, raw] = match
let value = raw
if (value.trim().startsWith('[') && !value.trim().endsWith(']')) {
while (i + 1 < end && !value.trim().endsWith(']')) value += ' ' + lines[++i].trim()
}
if (raw === '>' || raw === '|' || raw === '') {
const block: string[] = []
while (i + 1 < end && /^\s+\S/.test(lines[i + 1])) block.push(lines[++i].trim())
if (raw === '>') frontmatter[key] = block.join(' ')
else if (raw === '|') frontmatter[key] = block.join('\n')
else frontmatter[key] = block[0]?.startsWith('[') ? scalar(block.join(' ')) : block.length ? block.map(line => scalar(line.replace(/^-\s+/, ''))) : null
} else frontmatter[key] = scalar(value)
}
if (!Object.keys(frontmatter).length) throw new Error('Expected frontmatter fields between --- lines')
return { frontmatter, body: lines.slice(end + 1).join('\n').trim() }
export function parseRoster(text: string, path: string) {
const value = parse(text)
if (!value || !Array.isArray(value.roster)) throw new Error(`${path}: roster must be a list`)
const names = new Set<string>()
const definitions = value.roster.map((entry: any): Definition => {
if (!entry || typeof entry.name !== 'string' || !/^[a-zA-Z0-9_-]+$/.test(entry.name)) throw new Error(`${path}: invalid roster name`)
if (names.has(entry.name)) throw new Error(`${path}: duplicate agent ${entry.name}`)
names.add(entry.name)
for (const key of ['description', 'prompt', 'model']) if (typeof entry[key] !== 'string' || !entry[key].trim()) throw new Error(`${path}: ${entry.name}.${key} is required`)
if (entry.thinking !== undefined && !thinkingLevels.includes(entry.thinking)) throw new Error(`${path}: invalid thinking for ${entry.name}`)
for (const key of ['tools', 'addTools', 'denyTools']) if (entry[key] !== undefined && (!Array.isArray(entry[key]) || !entry[key].every((tool: any)=>typeof tool === 'string'))) throw new Error(`${path}: ${entry.name}.${key} must be a list of tool names`)
const {prompt, ...settings} = entry
const definition = {...settings, path, instructions:prompt}
definitionRuns(definition)
return definition
})
return {...value, definitions}
}
function settings(data: Record<string, unknown>, overlay: boolean) {
const unknown = Object.keys(data).find(key => !fields.includes(key) && !(overlay && key === 'addTools'))
if (unknown) throw new Error(`Unknown field: ${unknown}`)
if (typeof data.name !== 'string' || !/^[a-zA-Z0-9_-]+$/.test(data.name)) throw new Error('name must be a nonempty identifier')
for (const key of ['description', 'model']) {
if (data[key] !== undefined && (typeof data[key] !== 'string' || !data[key].trim())) throw new Error(key === 'description' ? 'description is required' : 'model must be a nonempty string')
}
if (data.family !== undefined && (typeof data.family !== 'string' || !familyName(data.family))) throw new Error('family must be a family name: lowercase letters, digits and -')
if (data.thinking !== undefined && !thinkingLevels.includes(data.thinking as Thinking)) throw new Error(`thinking must be one of ${thinkingLevels.join(', ')}`)
for (const key of ['tools', 'denyTools', 'addTools']) {
if (data[key] !== undefined && !tools(data[key])) throw new Error(`${key} must be a list of tool names`)
}
if (data.isolate !== undefined && data.isolate !== 'none' && data.isolate !== 'worktree') throw new Error('isolate must be none or worktree')
for (const key of ['resident', 'fast', 'summaries']) {
if (data[key] !== undefined && typeof data[key] !== 'boolean') throw new Error(`${key} must be a boolean`)
}
if (data.icon !== undefined && (typeof data.icon !== 'string' || !/^[A-Za-z][A-Za-z0-9]*(-[a-z0-9]+)*$/.test(data.icon))) throw new Error('icon must be a lucide icon name, like Gem or notebook-pen')
if (data.tint !== undefined && !['violet', 'blue', 'green', 'amber', 'red', 'pink', 'teal', 'grey'].includes(data.tint as string)) throw new Error('tint must be one of violet, blue, green, amber, red, pink, teal, grey')
return { ...data, ...(typeof data.description === 'string' ? { description: data.description.trim() } : {}) }
}
export function parseDefinition(text: string, path: string): Definition {
const { frontmatter, body } = parseFrontmatter(text)
const parsed = settings(frontmatter, false)
if (!parsed.description) throw new Error('description is required')
if (!body) throw new Error('Instructions are required')
return { ...parsed, path, instructions: body } as unknown as Definition
}
export function parseOverlay(text: string, path: string) {
const { frontmatter, body } = parseFrontmatter(text)
if ('tools' in frontmatter) throw new Error('An overlay adds tools with addTools, not tools')
const parsed = settings(frontmatter, true)
if (!tools(parsed.addTools)) throw new Error('addTools must be a list of tool names')
const { name, addTools, ...rest } = parsed
return { name: name as string, addTools, settings: rest as Partial<Definition>, path, instructions: body }
}
export function mergeOverlay(base: Definition, overlay: ReturnType<typeof parseOverlay>): Definition {
const { model: _pin, ...unpinned } = base
const suffix = overlay.settings.model?.match(/:(off|minimal|low|medium|high|xhigh|max)$/)
const resolved = suffix ? { ...overlay.settings, model: overlay.settings.model!.slice(0, -suffix[0].length), thinking: overlay.settings.thinking ?? suffix[1] as Thinking } : overlay.settings
return {
...(overlay.settings.family && !overlay.settings.model ? unpinned : base),
...resolved,
...(overlay.settings.denyTools ? { denyTools: [...new Set([...(base.denyTools ?? []), ...overlay.settings.denyTools])] } : {}),
...(base.tools ? { tools: [...new Set([...base.tools, ...overlay.addTools])] } : overlay.addTools.length ? { addTools: [...new Set([...(base.addTools ?? []), ...overlay.addTools])] } : {}),
instructions: overlay.instructions ? `${base.instructions}\n\n## Project additions\n\n${overlay.instructions}` : base.instructions,
overlays: [...(base.overlays ?? []), overlay.path],
}
}
export function parseFamilies(text: string): Families {
const value = JSON.parse(text)
if (!value || typeof value !== 'object' || Array.isArray(value)) throw new Error('families must be an object')
for (const [name, spec] of Object.entries(value) as [string, Family][]) {
if (!familyName(name) || !spec || typeof spec !== 'object' || !modelRef(spec.model)) throw new Error(`Invalid family: ${name}`)
if (spec.thinking !== undefined && !thinkingLevels.includes(spec.thinking)) throw new Error(`Invalid thinking for family: ${name}`)
}
if (!value.default) throw new Error('default family is required')
export function parseConfig(text: string) {
const value = parseRoster(text, CONFIG)
if (typeof value.prompt !== 'string') throw new Error(`${CONFIG}: prompt must be a string`)
if (!value.definitions.some((entry: Definition)=>entry.name === value.standing)) throw new Error(`${CONFIG}: standing must name a roster member`)
return value
}
function modelRef(model: unknown): model is string { return typeof model === 'string' && /^[^/\s]+\/\S+$/.test(model) }
export function definitionRuns(definition: Definition, families: Families, models: readonly string[]) {
const family = Object.hasOwn(families, definition.family ?? 'default') ? definition.family ?? 'default' : 'default'
let model = definition.model ?? families[family].model
let thinking = definition.thinking ?? (definition.model ? undefined : families[family].thinking)
const suffix = /:(off|minimal|low|medium|high|xhigh|max)$/.exec(model)
if (suffix) { model = model.slice(0, -suffix[0].length); thinking = definition.thinking ?? suffix[1] as Thinking }
if (!modelRef(model) || !models.includes(model)) throw new Error(`model: Model "${model}" not found`)
const slash = model.indexOf('/')
return { model: { provider: model.slice(0, slash), modelId: model.slice(slash + 1) }, thinking, ...(definition.model ? { via: 'pin' as const } : { via: 'family' as const, family }) }
export function definitionRuns(definition: Definition) {
if (!/^[^/\s]+\/\S+$/.test(definition.model)) throw new Error(`${definition.path}: invalid model for ${definition.name}`)
const slash = definition.model.indexOf('/')
return {model:{provider:definition.model.slice(0,slash),modelId:definition.model.slice(slash+1)},thinking:definition.thinking}
}
/** Sorted built-ins precede sorted project files. Invalid files produce errors without replacing an earlier definition. */
export function discoverAgents(files: readonly File[], builtinRoot: string, repo: string, families: Families, models: readonly string[]) {
const definitions = new Map<string, Definition & { source: 'builtin' | 'project' }>()
const errors: { name: string; path: string; source: string; error: string }[] = []
const replaced: { name: string; path: string; source: string }[] = []
for (const [source, root] of [['builtin', builtinRoot + '/agents/'], ['project', repo + '/.sausage/agents/']] as const) {
for (const file of files.filter(file => file.path.startsWith(root) && !file.path.slice(root.length).includes('/') && file.path.endsWith('.md')).toSorted((a, b) => a.path.localeCompare(b.path))) {
try {
let definition: Definition
if ('addTools' in parseFrontmatter(file.text).frontmatter) {
const overlay = parseOverlay(file.text, file.path)
const base = definitions.get(overlay.name)
if (!base) throw new Error(`addTools needs an earlier definition named ${overlay.name}`)
definition = mergeOverlay(base, overlay)
} else definition = parseDefinition(file.text, file.path)
definitionRuns(definition, families, models)
const previous = definitions.get(definition.name)
if (source === 'project' && previous?.source === 'builtin' && !definition.overlays) replaced.push({ name: definition.name, path: file.path, source })
definitions.set(definition.name, { ...definition, source: definition.overlays ? previous!.source : source })
} catch (error) { errors.push({ name: file.path.split('/').at(-1)!.slice(0, -3), path: file.path, source, error: String(error) }) }
}
}
return { definitions: [...definitions.values()].map(definition => ({ ...definition, runs: definitionRuns(definition, families, models) })), errors, replaced }
}
/** Tool denials also apply to inherited tools. Inputs stay unchanged. */
export function toolsFor(definition: Definition, parent: readonly string[]): string[] {
return [...new Set([...(definition.tools ?? parent), ...(definition.addTools ?? [])])].filter(tool => !definition.denyTools?.includes(tool))
}
extensions/dream/DREAM.md
+3 -3
View File
@@ -3,7 +3,7 @@
A dream is a read-only review in an ordinary branch. It does not change code, memory, settings, or storage automatically.
The user decides which proposals to accept. Approved proposals become ordinary work.
## For @ultimate
## For the standing agent
A user request or a built-in automation can request a dream.
If a dream branch is active, report that fact. Do not start another branch.
@@ -43,12 +43,12 @@ Do not edit files, apply proposals, or switch storage during a dream.
## Built-in automations
Dream creates two ordinary context automations after @ultimate exists:
Dream creates two ordinary context automations after the standing agent exists:
- `Dream before compaction`: the context exceeds 70%.
- `Dream after a pause`: the context exceeds 50% after eight hours without user input.
The automations wake @ultimate. Each fire also adds one passive `dream.requested` entry.
The automations wake the standing agent. Each fire also adds one passive `dream.requested` entry.
The automation ID and fire number identify that entry across replay and reopen.
Creation keys recover a lost receipt without a duplicate automation.
A user can pause or delete an automation. Reopen preserves those choices.
extensions/dream/built-ins.ts
+9 -8
View File
@@ -9,13 +9,14 @@ export const PAUSE_NAME='Dream after a pause'
export const BuiltInDoc=defineDoc<{id?:string;deleted?:true}>({kind:'dream.after-compaction',version:1,scope:'conversation',history:'latest',fork:'initial',checkpointWhen:()=>false,initial:()=>({})})
export const PauseDoc=defineDoc<{id?:string;deleted?:true}>({kind:'dream.after-pause',version:1,scope:'conversation',history:'latest',fork:'initial',checkpointWhen:()=>false,initial:()=>({})})
const builtIns=[
{doc:BuiltInDoc,name:AUTOMATION_NAME,condition:{metric:'context',of:'@ultimate',above:70}},
{doc:PauseDoc,name:PAUSE_NAME,condition:{metric:'context',of:'@ultimate',above:50,idleFor:8*3600_000}},
{doc:BuiltInDoc,name:AUTOMATION_NAME,condition:{metric:'context',above:70}},
{doc:PauseDoc,name:PAUSE_NAME,condition:{metric:'context',above:50,idleFor:8*3600_000}},
]
export const DREAM_ASK='Dream now: follow "For @ultimate" in /src/sausage/extensions/dream/DREAM.md. Write your handoff into the dream branch prompt in this turn, while your context still holds everything. If a dream branch is already active, say so and do not start another.'
export const DREAM_ASK='Dream now: follow "For the standing agent" in /src/sausage/extensions/dream/DREAM.md. Write your handoff into the dream branch prompt in this turn, while your context still holds everything. If a dream branch is already active, say so and do not start another.'
export async function ensureAutomation(host:any) {
const service=host.want(AutomationsApi)
if(!service || (await host.harness().snapshot(Standing,ROOT,context))?.ultimate===undefined) return
const standing=(await host.harness().snapshot(Standing,ROOT,context))?.['$standing']
if(!service || standing===undefined) return
for(const {doc,name,condition} of builtIns) {
const state=await host.harness().snapshot(doc,ROOT,context)
if(state?.deleted) continue
@@ -23,7 +24,7 @@ export async function ensureAutomation(host:any) {
const list=await service.operate({action:'list'})
if(!list.some((a:any)=>a.id===state.id)) await host.harness().commit(async(tx:any)=>{(await tx.doc(doc,ROOT)).deleted=true},context)
} else {
const made=await service.operate({action:'create',name,condition,subscribers:['@ultimate'],key:`dream:${doc.definition.kind}`})
const made=await service.operate({action:'create',name,condition:{...condition,of:`#${standing}`},subscribers:[standing],key:`dream:${doc.definition.kind}`})
await host.harness().commit(async(tx:any)=>{(await tx.doc(doc,ROOT)).id=made.id},context)
}
}
@@ -31,10 +32,10 @@ export async function ensureAutomation(host:any) {
export async function fired(host:any,tx:any,id:string,fire:number) {
const builtIn=await Promise.all(builtIns.map(async(item)=>({item,id:(await tx.doc(item.doc,ROOT)).id})))
const found=builtIn.find(item=>item.id===id)?.item
const ultimate=(await tx.doc(Standing,ROOT)).ultimate
if(!found || ultimate===undefined) return
const standing=(await tx.doc(Standing,ROOT))['$standing']
if(!found || standing===undefined) return
const automation=(await tx.doc(Automations)).automations[id]
const pause=found.doc===PauseDoc && automation ? Date.now()-(automation.condition?.inputAt??automation.createdAt) : undefined
const data={automation:{id,name:automation?.name??found.name,...(pause===undefined?{}:{pause})}}
await deliver(host,tx,ultimate,{type:'write',requestId:`dream.requested:${id}:${fire}`,kind:'dream.requested',data,content:`Automation ${id} "${data.automation.name}" fired${pause===undefined?'':` after ${Math.round(pause/3600_000)} h away`}. ${DREAM_ASK}`})
await deliver(host,tx,standing,{type:'write',requestId:`dream.requested:${id}:${fire}`,kind:'dream.requested',data,content:`Automation ${id} "${data.automation.name}" fired${pause===undefined?'':` after ${Math.round(pause/3600_000)} h away`}. ${DREAM_ASK}`})
}
extensions/dream/server.ts
+6 -6
View File
@@ -27,22 +27,22 @@ export default function dream(host:any) {
return run.catch((error:any)=>host.notice(`Dream automations: ${String(error)}`))
}
host.onOpen(ensure)
host.react('ultimate-started',[Started],(_tx:any,{data}:any)=>{
host.react('standing-started',[Started],(_tx:any,{data}:any)=>{
// Creation commits run after this reactor transaction completes.
if(data.kind==='standing'&&data.agent==='ultimate') void ensure()
if(data.kind==='standing') void ensure()
},{from:'now'})
host.react('requested',[Fired],(tx:any,{data}:any)=>fired(host,tx,data.id,data.fire),{from:'now'})
async function change(tx:any,args:any) {
const ultimate=(await tx.doc(Standing,ROOT)).ultimate
if(ultimate===undefined) throw new Error('@ultimate has no conversation yet')
const standing=(await tx.doc(Standing,ROOT))['$standing']
if(standing===undefined) throw new Error('the standing agent has no conversation yet')
if(args.action==='start') {
await deliver(host,tx,ultimate,{type:'input',requestId:`dream.requested:${crypto.randomUUID()}`,kind:'dream.requested',data:{},content:`The user requested a dream. ${DREAM_ASK}`})
await deliver(host,tx,standing,{type:'input',requestId:`dream.requested:${crypto.randomUUID()}`,kind:'dream.requested',data:{},content:`The user requested a dream. ${DREAM_ASK}`})
return {requested:true}
}
if(args.action!=='prepare') throw new Error(`Unknown dream action ${args.action}`)
const prepared=prepare(args,Date.now())
;(await tx.doc(Next,ROOT)).prepared=prepared
await deliver(host,tx,ultimate,{type:'write',requestId:`dream.prepared:${crypto.randomUUID()}`,kind:'dream.prepared',data:prepared,content:`Prepared dream handoff:\n${prepared.learned.join('\n')}\n\n${prepared.handoff}\nContinue in this world. No chapter switch is available.`})
await deliver(host,tx,standing,{type:'write',requestId:`dream.prepared:${crypto.randomUUID()}`,kind:'dream.prepared',data:prepared,content:`Prepared dream handoff:\n${prepared.learned.join('\n')}\n\n${prepared.handoff}\nContinue in this world. No chapter switch is available.`})
return prepared
}
host.implement(DreamApi,{operate:(args:any)=>host.harness().commit((tx:any)=>change(tx,args),context)})
extensions/goal/README.md
+3 -3
View File
@@ -1,15 +1,15 @@
# Hands-off goals
The user starts or ends a goal through the exported server actions.
Only the standing `@ultimate` conversation uses the `goal` tool.
Only the standing `the standing agent` conversation uses the `goal` tool.
The root document stores one open goal and its decisions.
Three failed tries block a criterion. Rewording preserves unchanged criteria.
Decisions can name screenshots from the caller's conversation.
The `latest` reference never crosses an input boundary.
Each input to `@ultimate` receives the current open goal note.
Each input to `the standing agent` receives the current open goal note.
A queued start card grants nothing after the user ends the goal.
The start wakes `@ultimate`. The report uses a passive write.
The start wakes `the standing agent`. The report uses a passive write.
The optional automations service creates one keyed reminder every 30 minutes.
A durable task synchronizes its state after each lifecycle change.
extensions/goal/delivery.ts
+1 -1
View File
@@ -3,7 +3,7 @@ import { defineTask } from '@earendil-works/pi-durable'
import type { ConversationId, Tx } from '@earendil-works/pi-durable'
import type { Host } from '../../host/js/extensions.ts'
/** A goal card for @ultimate: `goal.started` as an input that wakes it, `goal.report` as a write. */
/** A goal card for the standing agent: `goal.started` as an input that wakes it, `goal.report` as a write. */
type Card = { to: number; goal: number; kind: 'goal.started' | 'goal.report'; content: string }
/** Created in the commit that changes the goal, so a crash cannot lose the card; the request id makes retries harmless. */
extensions/goal/refs.ts
+10 -10
View File
@@ -1,7 +1,7 @@
/** Hands-off goals in the browser: the goal the user put @ultimate on, and every decision it took for them. Browser-safe. */
/** Hands-off goals in the browser: the goal the user put the standing agent on, and every decision it took for them. Browser-safe. */
import { actionRef, docRef } from '../goal/compat.ts'
/** One decision @ultimate took instead of asking the user. */
/** One decision the standing agent took instead of asking the user. */
export type Decision = {
at: number
/** What it would have asked the user. */
@@ -11,19 +11,19 @@ export type Decision = {
reason: string
/** The branch the decision was for. */
conversation?: number
/** The entry of @ultimate's conversation holding the screenshots it judged, such as a UI preview it approved. */
/** The entry of the standing agent's conversation holding the screenshots it judged, such as a UI preview it approved. */
images?: number
}
/** How a goal ended: met, blocked (every remaining criterion stuck), or ended by the user. */
export type Ending = {
at: number
by: 'user' | 'ultimate'
by: 'user' | 'standing'
outcome: 'met' | 'blocked' | 'ended'
summary?: string
}
/** What "done" means, as @ultimate wrote it down: each criterion open, met, or blocked after repeated failed tries. */
/** What "done" means, as the standing agent wrote it down: each criterion open, met, or blocked after repeated failed tries. */
export type Criterion = { text: string; status: 'open' | 'met' | 'blocked'; tries?: number }
export type Goal = {
@@ -33,9 +33,9 @@ export type Goal = {
startedAt: number
criteria: Criterion[]
decisions: Decision[]
/** The "Goal reminder" automation that wakes @ultimate every 30 minutes while the goal is open. */
/** The "Goal reminder" automation that wakes the standing agent every 30 minutes while the goal is open. */
reminder?: string
/** Set while @ultimate is blocked on the user: the reminder is paused until it resumes. */
/** Set while the standing agent is blocked on the user: the reminder is paused until it resumes. */
waiting?: { reason: string; at: number }
ended?: Ending
}
@@ -43,12 +43,12 @@ export type Goal = {
/** `{ next, goals }` on conversation 1, oldest first; at most one goal is open (no `ended`). */
export const GoalDoc = docRef<{ next: number; goals: Goal[] }>('goal')
/** `goal.started` and `goal.report` entries in @ultimate's conversation name their goal. */
/** `goal.started` and `goal.report` entries in the standing agent's conversation name their goal. */
export type GoalEntry = { goal: number }
/** `{ text }` → the new goal. User-only: puts @ultimate hands-off on it. */
/** `{ text }` → the new goal. User-only: puts the standing agent hands-off on it. */
export const StartAction = actionRef<{ text: string }, Goal>('goal', 'start')
/** No arguments → the ended goal. User-only: ends the open goal; @ultimate asks again as usual. */
/** No arguments → the ended goal. User-only: ends the open goal; the standing agent asks again as usual. */
export const EndAction = actionRef<null, Goal>('goal', 'end')
/** `{ goal, entry }` → the images of a decision's screenshot entry. */
export const ImagesAction = actionRef<
extensions/goal/reminder.ts
+4 -2
View File
@@ -1,5 +1,5 @@
/**
* The goal reminder: an automation that wakes @ultimate every 30 minutes while a goal is open, so a stalled goal never
* The goal reminder: an automation that wakes the standing agent every 30 minutes while a goal is open, so a stalled goal never
* waits silently. Every goal change (start, pause, resume, end) creates a sync task in its own commit; the task brings
* the automation in line with the goal as it is when the task runs. A crash leaves at most a task to rerun, and a rerun
* gets the automation an earlier try made back from its create key instead of making a second one.
@@ -9,6 +9,7 @@ import type { ConversationDocToken, Tx } from '@earendil-works/pi-durable'
import { BACKGROUND_CONTEXT as context } from '@earendil-works/chord/context'
import type { Host } from '../../host/js/extensions.ts'
import { AutomationsApi } from '../automations/server.ts'
import { Standing } from '../subagents/api.ts'
import type { Automation } from '../automations/types.ts'
import type { Goal } from './refs.ts'
@@ -54,11 +55,12 @@ export function reminderTask(host: Host, Goals: Goals) {
if (!own) {
// One the user deleted in the Automations panel stays deleted.
if (goal.reminder) return false
const standing = await host.harness().commit(async tx => (await tx.doc(Standing, ROOT_CONVERSATION_ID))['$standing'], context)
await automations.operate({
action: 'create',
name: reminderName(id),
schedule: SCHEDULE,
subscribers: ['@ultimate'],
subscribers: [standing],
key: reminderKey(id),
})
return true
extensions/goal/server.ts
+17 -17
View File
@@ -1,7 +1,7 @@
/**
* Hands-off goals: the user puts @ultimate on a goal and goes away. While it is open, @ultimate decides what it would
* Hands-off goals: the user puts the standing agent on a goal and goes away. While it is open, the standing agent decides what it would
* otherwise ask them, records each decision with its reason, and ends the goal with a report. Only the user starts or
* ends a goal early; @ultimate hears it on every run through a per-input note.
* ends a goal early; the standing agent hears it on every run through a per-input note.
*/
import { arr, int, lit, obj, oneOf, opt, str, union } from '@earendil-works/pi-schema'
import {
@@ -42,7 +42,7 @@ export const docs = [Goals]
/** A plain copy: the document's overlay is a proxy that structuredClone refuses, and unusable after the commit. */
const copy = (goal: Goal) => JSON.parse(JSON.stringify(goal)) as Goal
const ultimateOf = async (tx: Tx) => (await tx.doc(Standing, ROOT_CONVERSATION_ID)).ultimate
const standingOf = async (tx: Tx) => (await tx.doc(Standing, ROOT_CONVERSATION_ID))['$standing']
/** The open goal, if any. */
export async function openGoal(tx: Tx) {
@@ -87,7 +87,7 @@ const status = (criterion: Criterion) =>
: criterion.status
/**
* @ultimate's per-run note while a goal is open: the goal, its criteria, and how it works; on a fire of its reminder,
* the standing agent's per-run note while a goal is open: the goal, its criteria, and how it works; on a fire of its reminder,
* what to check.
*/
export function noteText(goal: Goal, reminded = false) {
@@ -135,8 +135,8 @@ type ToolArgs = {
summary?: string
}
/** One `goal` tool call by @ultimate (`ultimate`) in `tx`; returns its answer, and the goal when it finished. */
export async function act(tx: Tx, ultimate: ConversationId, args: ToolArgs, now: number) {
/** One `goal` tool call by the standing agent (`standing`) in `tx`; returns its answer, and the goal when it finished. */
export async function act(tx: Tx, standing: ConversationId, args: ToolArgs, now: number) {
const goal = await openGoal(tx)
if (!goal) throw new Error('No hands-off goal is open; ask the user as usual.')
const criterion = () => {
@@ -163,7 +163,7 @@ export async function act(tx: Tx, ultimate: ConversationId, args: ToolArgs, now:
...(args.conversation === undefined ? {} : { conversation: args.conversation }),
...(args.images === undefined
? {}
: { images: await imagesEntry(tx, ultimate, args.images) }),
: { images: await imagesEntry(tx, standing, args.images) }),
}
goal.decisions.push(decision)
return { answer: `Recorded decision ${goal.decisions.length}: ${decision.picked}` }
@@ -203,7 +203,7 @@ export async function act(tx: Tx, ultimate: ConversationId, args: ToolArgs, now:
throw new Error("finish needs outcome 'met' or 'blocked'")
const ended = await end(tx, {
at: now,
by: 'ultimate',
by: 'standing',
outcome: args.outcome,
summary: text(args.summary, 'summary'),
})
@@ -213,20 +213,20 @@ export async function act(tx: Tx, ultimate: ConversationId, args: ToolArgs, now:
}
}
/** The report card's model text: what @ultimate (or the user, ending it) should take from it. */
/** The report card's model text: what the standing agent (or the user, ending it) should take from it. */
const reportText = (goal: Goal) =>
goal.ended?.by === 'user'
? `The user ended hands-off goal ${goal.id}. From now on ask them as usual.`
: `Hands-off goal ${goal.id} ended (${goal.ended?.outcome}); its report card is here for the user. From now on ask them as usual.`
/**
* Goal `goal`'s card of `kind` for @ultimate, delivered by a task created in `tx`, the commit that changed the goal.
* The start card says only that the user started it: what @ultimate may do comes from the goal note, built from the
* Goal `goal`'s card of `kind` for the standing agent, delivered by a task created in `tx`, the commit that changed the goal.
* The start card says only that the user started it: what the standing agent may do comes from the goal note, built from the
* goal as it is when the input runs, so a start still queued after the user ended the goal grants nothing.
*/
async function card(host: Host, tx: Tx, kind: 'goal.started' | 'goal.report', goal: Goal) {
const to = await ultimateOf(tx)
if (to === undefined) throw new Error('@ultimate has no conversation yet')
const to = await standingOf(tx)
if (to === undefined) throw new Error('the standing agent has no conversation yet')
await deliver(host, tx, {
to,
goal: goal.id,
@@ -252,7 +252,7 @@ const sync = (host: Host, tx: Tx, goal: number) =>
},
)
/** Starting and ending a goal is the user's grant; @ultimate only finishes one through its tool. */
/** Starting and ending a goal is the user's grant; the standing agent only finishes one through its tool. */
export const userActions = ['start', 'end']
export const actions = {
start: implementAction(StartAction, (args, { host }) =>
@@ -299,7 +299,7 @@ export default function goalExtension(host: Host) {
// Pi upstream: @earendil-works/pi-durable@1.5.0 — patched transactional beforeInput hook.
// A per-run note while a goal is open, never a system-prompt section, keeps the prompt cached.
beforeInput: async (input, { tx, conversationId }) => {
if ((await ultimateOf(tx)) !== conversationId) return undefined
if ((await standingOf(tx)) !== conversationId) return undefined
const goal = await openGoal(tx)
if (!goal) return undefined
// World automations use a plain input with their fire prefix.
@@ -351,8 +351,8 @@ export default function goalExtension(host: Host) {
async execute(args, api, ctx) {
const caller = api.conversationId
const result = await api.commit(async (tx) => {
if ((await ultimateOf(tx)) !== caller)
throw new Error('Only @ultimate keeps the hands-off goal.')
if ((await standingOf(tx)) !== caller)
throw new Error('Only the standing agent keeps the hands-off goal.')
const acted = await act(tx, caller, args, Date.now())
if (acted.ended) await card(host, tx, 'goal.report', acted.ended)
const changed = acted.sync ?? acted.ended?.id
extensions/lineage/server.ts
+3 -3
View File
@@ -772,7 +772,7 @@ async function refuseTally(tx: Tx, id: ConversationId) {
if (!(await tx.doc(Roots, ROOT_CONVERSATION_ID)).roots.includes(id)) return
const tally = await tx.doc(Tally, id)
if (tally.since !== undefined)
throw new Error(`#${id} is a tally and runs no turns; ask @ultimate to start a branch`)
throw new Error(`#${id} is a tally and runs no turns; ask the standing agent to start a branch`)
}
@@ -964,9 +964,9 @@ export default function lineageExtension(host: any) {
async execute({ root, ops }, api, ctx) {
// Every conversation sees the tool, but only the primary agent can write memory.
const standing = await host.harness().snapshot(Standing, ROOT_CONVERSATION_ID, ctx)
if (standing?.ultimate !== api.conversationId)
if (standing?.['$standing'] !== api.conversationId)
throw new Error(
'Only @ultimate writes project memory; tell @ultimate what to remember (message, reply: false).',
'Only the standing agent writes project memory; tell the standing agent what to remember (message, reply: false).',
)
const id = await memoryRoot(api.conversationId, root, ctx)
const summary = await applyMemoryOps(host, id, ops, Date.now(), `tool:${api.taskId}:${api.callId}`)
extensions/messages/README.md
+3 -3
View File
@@ -1,6 +1,6 @@
# Messages and lineage
The extensions preserve the durable `board.*` and `lineage.*` identifiers from Ultimate.
The extensions preserve the durable `board.*` and `lineage.*` identifiers from Standing agent.
They use world files and world git. They do not use native filesystem or process APIs.
## Branches and delivery
@@ -27,7 +27,7 @@ The starter's final answer reaches the branch without a faulted routing task.
A static test checks free runtime identifiers in every extension module, including function bodies.
The grant chain contains fork ancestry before starter ancestry.
Each conversation occurs once. A root fork that @ultimate starts includes `[branch, root, ultimate]`.
Each conversation occurs once. A root fork that the standing agent starts includes `[branch, root, standing]`.
## Tally and catch-up
@@ -54,7 +54,7 @@ A shared queue serializes writes across extension reloads.
A tool or landing receipt commits in that same file.
Thus, a restart after the rename does not apply the operations twice.
Only @ultimate can use the `remember` tool.
Only the standing agent can use the `remember` tool.
Other extensions can use the lineage service.
Branches from a root with memory start with its index and recent landings.
extensions/messages/addresses.ts
+2 -2
View File
@@ -36,9 +36,9 @@ export async function parentOf(host: ExtensionHost, id: number, ctx: Context) {
}
export async function senderName(host: ExtensionHost, id: number | 'user', ctx: Context) {
if (id === 'user') return 'user'
if (typeof id !== 'number') return 'another Ultimate'
if (typeof id !== 'number') return 'another Standing agent'
const index = await host.harness().snapshot(Standing, ROOT_CONVERSATION_ID, ctx)
const name = Object.keys(index ?? {}).find((candidate) => index?.[candidate] === id)
const name = Object.keys(index ?? {}).find((candidate) => candidate !== '$standing' && index?.[candidate] === id)
return `#${id}${name ? ` (@${name})` : ''}`
}
export async function resolve(
extensions/messages/api.ts
+1 -1
View File
@@ -37,7 +37,7 @@ export type Message = {
echo?: true
/** The questions a reply-expected message to the user asks, answered on its card. */
questions?: Question[]
/** A user answer or dismissal that @ultimate settled for the user, not their own click. */
/** A user answer or dismissal that the standing agent settled for the user, not their own click. */
via?: Via
}
/** Every message between conversations and the user, on the root conversation. */
extensions/messages/delivery.ts
+1 -1
View File
@@ -34,7 +34,7 @@ export function delivery(host: ExtensionHost) {
note: boolean
fromLabel: string
toLabel: string
/** Another project's Ultimate answered `message`: deliver only this reply. */
/** Another project's Standing agent answered `message`: deliver only this reply. */
reply?: { message: Message; failed: boolean; stopped?: boolean }
},
| { phase: 'deliver' }
extensions/messages/notice.ts
+2 -2
View File
@@ -1,4 +1,4 @@
/** Passive context for @ultimate about a branch's questions to the user. */
/** Passive context for the standing agent about a branch's questions to the user. */
import { defineTask } from '@earendil-works/pi-durable'
import type { ConversationId } from '@earendil-works/pi-durable'
import type { Host as ExtensionHost } from '../../host/js/extensions.ts'
@@ -14,7 +14,7 @@ export const notice = (host: Pick<ExtensionHost, 'harness'>) =>
call?: string
summary?: string
dismissed?: boolean
/** @ultimate settled it for the user. */
/** the standing agent settled it for the user. */
via?: { conversation: number; entry?: number }
}
content: string
extensions/messages/questions.ts
+3 -3
View File
@@ -129,8 +129,8 @@ export const gist = (message: Pick<Message, 'text' | 'questions'>) => {
* open, so the agent treats them as unanswered. A per-input entry, not a system-prompt section, keeps the prompt cached.
*/
export async function openQuestionsNote(tx: Tx, conversationId: ConversationId, now: number) {
// @ultimate sees every open question to the user, since it settles those the user decides in its chat.
const primary = (await tx.doc(Standing, ROOT_CONVERSATION_ID)).ultimate === conversationId
// the standing agent sees every open question to the user, since it settles those the user decides in its chat.
const primary = (await tx.doc(Standing, ROOT_CONVERSATION_ID))['$standing'] === conversationId
const open = (await tx.doc(Messages, ROOT_CONVERSATION_ID)).messages.filter(
(m) => openQuestion(m) && (primary || m.from === conversationId),
)
@@ -143,7 +143,7 @@ export async function openQuestionsNote(tx: Tx, conversationId: ConversationId,
)
.join(
'\n',
)}${primary ? `\nWhen the user decides one in your chat, settle it with ultimate.action('messages', 'settle', { messageId, text }).` : ''}`
)}${primary ? `\nWhen the user decides one in your chat, settle it with standing.action('messages', 'settle', { messageId, text }).` : ''}`
return {
entries: [
{
extensions/messages/route.ts
+1 -1
View File
@@ -46,7 +46,7 @@ async function live(tx: Tx, kind: string, conversationId?: ConversationId) {
*/
async function busy(tx: Tx, id: ConversationId) {
if ((await tx.doc(LiveDoc, id)).run || (await tx.doc(InboxDoc, id)).items.length) return true
if ((await live(tx, 'ultimate.subagent-reporter', id)).length) return true
if ((await live(tx, 'standing.subagent-reporter', id)).length) return true
return (await live(tx, 'lineage.notify', id)).some(
(task) => (task.input as { wake?: unknown }).wake !== false,
)
extensions/messages/send.ts
+5 -5
View File
@@ -71,7 +71,7 @@ export async function send(
/** Place it as a note, which starts no run; by default one-way messages to the sender's starter are notes. */
note?: boolean
device?: Device
/** The user's answer or dismissal settled by @ultimate on their behalf, from their input `via.entry`. */
/** The user's answer or dismissal settled by the standing agent on their behalf, from their input `via.entry`. */
via?: Via
/** A branch's final answer routed as its report: the branch shows as awaiting its starter. */
report?: boolean
@@ -91,7 +91,7 @@ export async function send(
if (callId !== undefined && sent?.ids[callId] !== undefined) return sent.ids[callId]
}
const standing = await host.harness().snapshot(Standing, ROOT_CONVERSATION_ID, ctx)
const primaryId = standing?.ultimate
const primaryId = standing?.['$standing']
// The primary agent's messages to a busy conversation almost always steer it: new scope, a decision, a stop.
const steer = input.steer ?? (from !== 'user' && from === primaryId)
// Checked before resolving, which may create a standing agent's conversation.
@@ -100,7 +100,7 @@ export async function send(
if (input.to === 'user') {
if (from !== primaryId && !(parent === undefined && starter !== undefined))
throw new Error(
'Only @ultimate and branches message the user; send this to the conversation that started you.',
'Only the standing agent and branches message the user; send this to the conversation that started you.',
)
}
if (!input.text.trim()) throw new Error('Message text must not be empty')
@@ -113,7 +113,7 @@ export async function send(
)
const to = await resolve(host, input.to, from, ctx)
const fromLabel = via
? `user (settled for them by @ultimate, from what they said in its chat)`
? `user (settled for them by the standing agent, from what they said in its chat)`
: await senderName(host, from, ctx)
const toLabel = await senderName(host, to, ctx)
return host.harness().commit(async (tx) => {
@@ -261,7 +261,7 @@ export async function send(
)
}
// The user's answer or dismissal is their input to the asker; the message's `entry` names it once delivered. One
// @ultimate settled for them is not: their chat input that it came from counted already.
// the standing agent settled for them is not: their chat input that it came from counted already.
if (from === 'user' && answerId !== undefined && typeof to === 'number' && !via)
await host.emit(tx, UserInputEvent, {
conversation: to,
extensions/messages/server.ts
+7 -7
View File
@@ -26,7 +26,7 @@ async function openMessage(host: ExtensionHost, messageId: number, ctx: Context)
if (!message || !openQuestion(message)) throw new Error('These questions are no longer open')
return message
}
/** The user's answer `text` to open message `messageId`, or @ultimate's on their behalf (`via`). */
/** The user's answer `text` to open message `messageId`, or the standing agent's on their behalf (`via`). */
async function answer(
host: ExtensionHost,
messageId: number,
@@ -35,7 +35,7 @@ async function answer(
options: { device?: Device; via?: { via: Via; primary: number } } = {},
) {
const message = await openMessage(host, messageId, ctx)
// @ultimate settling its own question has the answer already: a one-way note, not another turn.
// the standing agent settling its own question has the answer already: a one-way note, not another turn.
const own = options.via !== undefined && message.from === options.via.primary
return send(
host,
@@ -50,7 +50,7 @@ async function answer(
},
)
}
/** Close open message `messageId` unanswered: the user's dismissal, or @ultimate's on their behalf with `reason`. */
/** Close open message `messageId` unanswered: the user's dismissal, or the standing agent's on their behalf with `reason`. */
async function dismiss(
host: ExtensionHost,
messageId: number,
@@ -106,7 +106,7 @@ export default function messages(host: ExtensionHost) {
let by = (await api.snapshot(Started, api.conversationId, ctx))?.by
const down =
by === undefined &&
(await api.snapshot(Standing, ROOT_CONVERSATION_ID, ctx))?.ultimate ===
(await api.snapshot(Standing, ROOT_CONVERSATION_ID, ctx))?.['$standing'] ===
api.conversationId
if (down) by = api.conversationId
if (by === undefined) return undefined
@@ -146,7 +146,7 @@ export default function messages(host: ExtensionHost) {
starter === undefined
? ''
: ` #${starter} started you: when you stop with nothing under way (no subagent, landing or reply pending), your final answer goes to it and wakes it; while work is under way it arrives as a note, and so do your one-way messages to it.`
return `Use message to contact #id or @name. @ultimate and branches may message user. Messages wake their recipient; its answer returns as a reply and wakes you. End your turn instead of waiting. A one-way message (reply: false) gets no answer; use reply: true when you need one. Inputs identify their sender; answer here and your answer is sent to the sender automatically; do not use message to reply. A message to an agent whose message you still owe an answer becomes that answer and ends your run; use it only to attach images.${upward}\n${agents.map((a) => `@${a.name} — ${a.description}`).join('\n')}`
return `Use message to contact #id or @name. the standing agent and branches may message user. Messages wake their recipient; its answer returns as a reply and wakes you. End your turn instead of waiting. A one-way message (reply: false) gets no answer; use reply: true when you need one. Inputs identify their sender; answer here and your answer is sent to the sender automatically; do not use message to reply. A message to an agent whose message you still owe an answer becomes that answer and ends your run; use it only to attach images.${upward}\n${agents.map((a) => `@${a.name} — ${a.description}`).join('\n')}`
}),
],
tools: [
@@ -154,7 +154,7 @@ export default function messages(host: ExtensionHost) {
defineTool({
name: 'message',
description:
'Send a message to #id or @name; it returns immediately. A busy recipient gets it after its current run, or during it with steer: true (the default for @ultimate). images forwards screenshots or other images from your conversation, to an agent or to the user. To an agent or the user whose message you owe an answer, it is that answer (with its images and questions) and ends your run.',
'Send a message to #id or @name; it returns immediately. A busy recipient gets it after its current run, or during it with steer: true (the default for the standing agent). images forwards screenshots or other images from your conversation, to an agent or to the user. To an agent or the user whose message you owe an answer, it is that answer (with its images and questions) and ends your run.',
parameters: obj({
to: str(),
text: str({ minLength: 1 }),
@@ -196,7 +196,7 @@ export default function messages(host: ExtensionHost) {
}
if (input.to !== 'user')
return { output: [{ type: 'text', text: `Message ${id} sent.` }] }
// The message is @ultimate's answer: the run ends here instead of restating it, unless a reply it has read
// The message is the standing agent's answer: the run ends here instead of restating it, unless a reply it has read
// is owed. One still queued gets its own turn.
const owed = await host
.harness()
extensions/models/README.md
+13 -10
View File
@@ -1,21 +1,22 @@
# Model selection
The `models` service starts a conversation with `start({agent, cwd, ownership?}, context)`.
The `agent` field names a built-in definition or a project overlay.
The `agent` field names a roster entry in `/etc/sausage/config.yaml`.
The service stores that name in `models.profile`.
`discover(cwd)` exposes the same live definition resolver to subagents.
`configure(tx, id, {agent, inheritedTools, excludeTools?, model?, thinking?})` configures an existing child in its creation transaction.
It retains the parent's tool scope, explicit overrides, and exclusions for subsequent run boundaries.
It applies the definition during creation and before each new run.
Definitions without an explicit model follow their family, then the default family.
Each roster entry names its model and thinking level directly. There are no model families or project overlays.
Each resolution reads `/src/sausage/agents/models.json` and the definition files through `host.fs`.
Each resolution reads `/etc/sausage/config.yaml` through `host.fs`.
A file edit applies at the next run boundary without a restart.
The project definitions live under `<repo>/.sausage/agents/`.
The existing T37 parser controls validation, overlays, model suffixes, and tool denials.
The global `prompt` precedes the role-specific `prompt` in every agent's system prompt.
A standing conversation follows the current `standing` roster member, even after a rename.
The optional `models.pin` service supplies `pin(tx, conversationId)` for an external model override.
Without that provider, the definition or default family selects the model.
Without that provider, the roster entry selects the model.
A conversation without a role uses the standing roster member's model.
No login fallback runs.
Existing registered models retain their transport.
@@ -30,7 +31,9 @@ The extension never reads a credential or registers a direct provider SDK.
OpenAI models use pi-ai's Responses transport. Anthropic models use its Messages transport.
The catalog endpoint determines each model's base URL.
For Anthropic, effort capabilities select adaptive thinking and its advertised effort levels.
Both providers use the existing `llm-proxy` key and usage tag. No provider login or direct endpoint fallback runs.
Both providers read the proxy key and base URL from `/etc/sausage/config.yaml` before each request.
The usage tag remains unchanged. No provider login or direct endpoint fallback runs.
Empty proxy settings do not prevent startup. A model request names the config file if either setting is empty.
If the catalog fails or exceeds five seconds, a small table supplies these defaults:
@@ -43,7 +46,7 @@ Catalog prices use USD per million tokens, with conversion from `unit_tokens` wh
The host uses the first advertised route's quote, including cache reads and cache writes.
These costs are estimates, not billing totals. Another proxy route can charge different prices.
Unavailable prices remain zero. Catalog changes apply after a host restart.
World family edits still apply at the next run boundary without a restart.
`host/tests/proxy_models.rs` tests the real transports against a local fake proxy.
Roster edits still apply at the next run boundary without a restart.
`host/tests/it/proxy_models.rs` tests the real transports against a local fake proxy.
`host/tests/core_ext.rs` tests these services through the snapshot loader.
`host/tests/it/core_ext.rs` tests these services through the snapshot loader.
extensions/models/server.ts
+12 -27
View File
@@ -1,32 +1,15 @@
/** Resolve world families and definitions at each run boundary. All real requests retain the host proxy transport. */
/** Read the world config at every run boundary. All real requests retain the host proxy transport. */
import { AgentDoc, defineExtension, GenerationTask, hook } from '@earendil-works/pi-durable'
import { definitionRuns, discoverAgents, parseDefinition, parseOverlay, parseFamilies, parseFrontmatter, toolsFor } from '../core/agents.ts'
import { CONFIG, definitionRuns, parseConfig, toolsFor } from '../core/agents.ts'
import { ModelsApi, Pin, Profile } from './api.ts'
import { repoRoot } from './files.ts'
export const order = -200
export default function models(host: any) {
const read = async (cwd: string) => {
const families = parseFamilies(await host.fs.read('/src/sausage/agents/models.json'))
const root = await repoRoot(host.fs, cwd)
const files = []
for (const folder of ['/src/sausage/agents', `${root}/.sausage/agents`]) {
let entries
try { entries = await host.fs.list(folder) } catch (error) { if (error.code !== 'not_found') throw error; entries = [] }
for (const entry of entries) if (entry.kind === 'file' && entry.name.endsWith('.md')) {
const path = `${folder}/${entry.name}`
files.push({path, text:await host.fs.read(path)})
}
}
const refs = Object.values(families).map(spec => spec.model.replace(/:(off|minimal|low|medium|high|xhigh|max)$/, ''))
// Explicit pins can name a proxy model outside the families.
for (const file of files) {
try {
const parsed = 'addTools' in parseFrontmatter(file.text).frontmatter ? parseOverlay(file.text,file.path).settings : parseDefinition(file.text,file.path)
if (parsed.model) refs.push(parsed.model.replace(/:(off|minimal|low|medium|high|xhigh|max)$/, ''))
} catch { /* discoverAgents records invalid files without replacing valid definitions. */ }
}
const found = discoverAgents(files, '/src/sausage', root, families, refs)
return {families, ...found}
const read = async (_cwd: string) => {
const config = parseConfig(await host.fs.read(CONFIG))
return {...config, definitions:config.definitions.map((definition: any)=>({...definition,
resident:definition.name === config.standing || definition.resident,
runs:definitionRuns(definition)})), errors:[], replaced:[]}
}
const register = (ref: any) => {
const store = host.models()
@@ -42,17 +25,18 @@ export default function models(host: any) {
const state = await tx.doc(AgentDoc, id)
const profile = await tx.doc(Profile, id)
const config = await read(state.cwd ?? '/')
if (profile.standing) profile.agent = config.standing
const definition = profile.agent ? config.definitions.find(d => d.name === profile.agent) : undefined
if (profile.agent && !definition) throw new Error(`Unknown agent ${profile.agent}: ${JSON.stringify(config.errors)}`)
const selected = definition?.runs ?? definitionRuns({} as any,config.families,Object.values(config.families).map(spec=>spec.model.replace(/:(off|minimal|low|medium|high|xhigh|max)$/, '')))
const selected = (definition ?? config.definitions.find((d: any)=>d.name === config.standing)).runs
const pinned = await host.want(Pin)?.pin(tx, id)
const ref = pinned ?? profile.model ?? selected.model
register(ref)
state.model = ref
if (profile.thinking ?? selected.thinking) state.thinkingLevel = profile.thinking ?? selected.thinking
else delete state.thinkingLevel
state.instructions = [config.prompt, definition ? `You are @${definition.name}.\n\n${definition.instructions}` : ''].filter(Boolean).join('\n\n')
if (definition) {
state.instructions = definition.instructions
if (profile.inheritedTools === undefined) profile.inheritedTools = state.tools ?? null
const inherited = profile.inheritedTools
if (definition.tools || Array.isArray(inherited)) state.tools = toolsFor(definition,inherited ?? []).filter(name=>!profile.excludeTools?.includes(name))
@@ -65,6 +49,7 @@ export default function models(host: any) {
configure: async (tx: any, id: number, options: any) => {
const profile = await tx.doc(Profile,id)
profile.agent = options.agent
profile.standing = options.standing ?? false
profile.inheritedTools = options.inheritedTools
profile.excludeTools = options.excludeTools ?? []
if (options.model) profile.model = options.model
extensions/pause/README.md
+1 -1
View File
@@ -23,4 +23,4 @@ A manual pause always requires Play.
The timer operates outside the held scheduler.
The stored pause re-arms the timer after restart.
The test module ports all 19 Ultimate pause cases through snapshot-loaded modules.
The test module ports all 19 Standing agent pause cases through snapshot-loaded modules.
extensions/pause/api.ts
+1 -1
View File
@@ -1,4 +1,4 @@
/** Pause: whether the user paused this Ultimate, for timers that start work outside the task scheduler. */
/** Pause: whether the user paused this Standing agent, for timers that start work outside the task scheduler. */
const defineApi = <S>() => (name: string): {name: string; service?: S} => ({name})
export type PauseService = {
extensions/pause/refs.ts
+8 -8
View File
@@ -1,10 +1,10 @@
/** Pause in the browser: whether this Ultimate is paused, what still runs, and Pause and Play. Browser-safe. */
/** Pause in the browser: whether this Standing agent is paused, what still runs, and Pause and Play. Browser-safe. */
const actionRef = <A, R>(extension: string, name: string) => ({extension, name})
const docRef = <S>(kind: string) => ({kind})
/**
* The rate limit that paused Ultimate: the provider and model that answered with it, the error as Pi recorded it,
* when (ms), and the conversation it first hit. Also the data of the `pause.limit` card in @ultimate's transcript.
* The rate limit that paused Standing agent: the provider and model that answered with it, the error as Pi recorded it,
* when (ms), and the conversation it first hit. Also the data of the `pause.limit` card in the standing agent's transcript.
*/
export type RateLimit = {
provider: string
@@ -25,7 +25,7 @@ export type PauseDocState = {
limit?: RateLimit
/** The user pressed Pause during this pause: it never plays by itself. */
manual?: boolean
/** The user's setting: minutes after a rate-limit pause began that Ultimate presses Play by itself; unset is off. */
/** The user's setting: minutes after a rate-limit pause began that Standing agent presses Play by itself; unset is off. */
autoPlay?: number
}
@@ -44,7 +44,7 @@ export type RunningTask = {
export type PauseState = {
paused: boolean
since?: number
/** The rate limit that paused Ultimate, if one did. */
/** The rate limit that paused Standing agent, if one did. */
limit?: RateLimit
/** Minutes after a rate-limit pause began that Play is pressed by itself; unset is off. */
autoPlay?: number
@@ -66,7 +66,7 @@ export const PauseAction = actionRef<null, PauseState>('pause', 'pause')
/** User-only: continue. Each run that was mid-turn when paused gets one "Paused for …" note. Returns the state. */
export const PlayAction = actionRef<null, PauseState>('pause', 'play')
/**
* User-only `{ minutes }`: play by itself that many minutes (1 to 1440) after a rate limit paused Ultimate, once per
* User-only `{ minutes }`: play by itself that many minutes (1 to 1440) after a rate limit paused Standing agent, once per
* pause; `null` turns it off. A pause the user pressed never plays by itself. Returns the state.
*/
export const AutoPlayAction = actionRef<{ minutes: number | null }, PauseState>('pause', 'autoPlay')
@@ -79,9 +79,9 @@ export type PauseNote = { since: number; until: number; finished?: boolean; auto
/**
* The `pause.held` entry in a subagent's parent, mid-turn when the subagent hit a rate limit: subagent `child` hit
* `provider`'s limit for `model` at `at` (ms), and Ultimate is paused until Play.
* `provider`'s limit for `model` at `at` (ms), and Standing agent is paused until Play.
*/
export type HeldNote = { child: number; provider: string; model: string; at: number }
/** The `pause.autoplay` entry in @ultimate's transcript: the rate-limit pause from `since` played by itself at `at`. */
/** The `pause.autoplay` entry in the standing agent's transcript: the rate-limit pause from `since` played by itself at `at`. */
export type AutoPlayed = { since: number; at: number; minutes: number }
extensions/pause/server.ts
+15 -15
View File
@@ -63,9 +63,9 @@ function serial<T>(host: Pick<ExtensionHost, 'resource'>, work: () => Promise<T>
* while paused and execute nothing meanwhile, so they never keep the pause from being safe.
*/
const PASSIVE: Record<string, string[]> = {
'ultimate.subagent-reporter': ['deliver', 'retry'],
'ultimate.automation-scheduler': ['sleep'],
'ultimate.automation-run': ['deliver'],
'standing.subagent-reporter': ['deliver', 'retry'],
'standing.automation-scheduler': ['sleep'],
'standing.automation-run': ['deliver'],
'board.delivery': ['deliver', 'reply'],
'board.route': ['observe'],
'conversations.summary': ['observe'],
@@ -213,7 +213,7 @@ function arm(host: Host, doc: PauseDocState) {
/**
* Play the pause by itself once its time came (`now`, for tests), once: Play ends the pause, and a limit that still
* holds pauses again with a new start and a full grace period. Tells @ultimate with a `pause.autoplay` entry first,
* holds pauses again with a new start and a full grace period. Tells the standing agent with a `pause.autoplay` entry first,
* while the pause is still recorded: a crash before Play reopens paused, re-arms the timer, and the entry's request ID
* keeps the second announcement from repeating.
*/
@@ -224,10 +224,10 @@ export async function autoPlay(host: Host, now = Date.now()) {
if (at === undefined || now < at) return false
const data: AutoPlayed = { since: doc.since!, at: now, minutes: doc.autoPlay! }
const harness = host.harness()
const ultimate = (await harness.snapshot(Standing, ROOT_CONVERSATION_ID, context))?.ultimate
if (ultimate !== undefined)
const standing = (await harness.snapshot(Standing, ROOT_CONVERSATION_ID, context))?.['$standing']
if (standing !== undefined)
await (
await harness.conversation(ultimate, context)
await harness.conversation(standing, context)
)?.submit(
{
type: 'write',
@@ -238,7 +238,7 @@ export async function autoPlay(host: Host, now = Date.now()) {
model: [
{
role: 'user',
content: `Ultimate played by itself ${lasted(now - data.since)} after the rate limit paused it (the user's setting: ${data.minutes} min).`,
content: `Standing agent played by itself ${lasted(now - data.since)} after the rate limit paused it (the user's setting: ${data.minutes} min).`,
timestamp: now,
},
],
@@ -251,7 +251,7 @@ export async function autoPlay(host: Host, now = Date.now()) {
})
}
/** The user's setting: play by itself `minutes` after a rate limit paused Ultimate; `null` turns it off. */
/** The user's setting: play by itself `minutes` after a rate limit paused Standing agent; `null` turns it off. */
async function setAutoPlay(host: Host, minutes: unknown) {
if (
minutes !== null &&
@@ -268,7 +268,7 @@ async function setAutoPlay(host: Host, minutes: unknown) {
}
/**
* A model answered with a rate limit: pause, so nothing retries into it, and tell @ultimate with one `pause.limit`
* A model answered with a rate limit: pause, so nothing retries into it, and tell the standing agent with one `pause.limit`
* card. Pi has already scheduled that generation's retry; the hold parks it, and every other run at its next step,
* until Play. The caller holds the scheduler already, in the commit's publication, before that retry's timer can fire.
*/
@@ -318,12 +318,12 @@ async function held(host: Host, limit: RateLimit) {
/** The `pause.limit` card for the limit that paused the host, once per pause (its request ID is the pause's start). */
async function card(host: Host, doc: PauseDocState) {
const harness = host.harness()
const ultimate = (await harness.snapshot(Standing, ROOT_CONVERSATION_ID, context))?.ultimate
if (!doc.paused || !doc.limit || ultimate === undefined) return
// Only kind and data, so Pi appends it at once even while @ultimate's own run is held; the model learns of the limit
const standing = (await harness.snapshot(Standing, ROOT_CONVERSATION_ID, context))?.['$standing']
if (!doc.paused || !doc.limit || standing === undefined) return
// Only kind and data, so Pi appends it at once even while the standing agent's own run is held; the model learns of the limit
// from the note Play leaves on each held run.
await (
await harness.conversation(ultimate, context)
await harness.conversation(standing, context)
)?.submit(
{
type: 'write',
@@ -380,7 +380,7 @@ async function play(host: Host, auto = false) {
const content =
how === 'finished'
? `Paused for ${span} after this run's last answer.`
: `Paused for ${span}; the run continued after ${auto ? 'Ultimate played by itself' : 'Play'}.`
: `Paused for ${span}; the run continued after ${auto ? 'Standing agent played by itself' : 'Play'}.`
const conversation = await harness.conversation(id, context)
await conversation?.submit(
{
extensions/plan/README.md
+2 -2
View File
@@ -1,6 +1,6 @@
# Plan
The `plan` tool maintains a durable epic/task ledger in `ultimate.plan`.
The `plan` tool maintains a durable epic/task ledger in `standing.plan`.
Task-owned descendants read their top owner's ledger. Only that owner edits it.
Ownerless conversations maintain separate ledgers.
@@ -8,7 +8,7 @@ Edits and receipts commit together. Memo namespaces distinguish nested calls tha
The tool returns both JSON text and structured content.
The compact prompt section excludes completed epics.
The root document `ultimate.plan.tasks` indexes linked child conversations.
The root document `standing.plan.tasks` indexes linked child conversations.
A new link or task title change updates the child's metadata title.
Other edits preserve a title that the user supplied.
extensions/plan/server.ts
+3 -3
View File
@@ -37,7 +37,7 @@ export type Epic = {
tasks: Record<string, Task>
}
export const Plan = defineDoc<{ next: number; epics: Record<string, Epic> }>({
kind: 'ultimate.plan',
kind: 'standing.plan',
checkpointWhen: () => false,
version: 1,
scope: 'conversation',
@@ -54,7 +54,7 @@ export type TaskLink = {
status: Task['status']
}
export const TaskIndex = defineDoc<{ children: Record<string, TaskLink> }>({
kind: 'ultimate.plan.tasks',
kind: 'standing.plan.tasks',
checkpointWhen: () => false,
version: 1,
scope: 'conversation',
@@ -65,7 +65,7 @@ export const TaskIndex = defineDoc<{ children: Record<string, TaskLink> }>({
export const docs = [Plan, TaskIndex]
// Commit the receipt with the edit, closing the crash window before api.memo stores the result.
const Receipt = defineDoc<{ results: Record<string, string> }>({
kind: 'ultimate.plan.receipt',
kind: 'standing.plan.receipt',
checkpointWhen: () => false,
version: 1,
scope: 'task',
extensions/prompt/README.md
+9 -3
View File
@@ -1,6 +1,12 @@
# System context
The prompt extension supplies `project_context`, `skills`, `tools`, and `cwd` sections.
Every agent receives the top-level `prompt` from `/etc/sausage/config.yaml`, followed by its roster entry's `prompt`.
The models extension reads both prompts again before each run.
An edit applies to an existing conversation at its next run without a host restart.
The default global prompt explains the world, its source, its config, and the harness routes.
Agents can edit that shared guidance through the config.
The prompt extension supplies only runtime context: `project_context`, `skills`, `tools`, and `cwd` sections.
The effective environment cwd takes precedence over the configured agent cwd.
Project instructions include each AGENTS.md from the repository root to that cwd, in that order.
Outside a repository, the extension reads the cwd's AGENTS.md only.
@@ -9,6 +15,6 @@ Skills come from `/src/sausage/skills/*/SKILL.md` and `<repo>/.sausage/skills/*/
The prompt lists each skill's name, description, and path.
A project skill replaces a built-in skill with the same name.
The tools section lists the tools that the resolved agent can use.
No Ultimate build note enters the prompt.
`host/tests/core_ext.rs` tests these sections through the real snapshot loader.
`host/tests/it/core_ext.rs` tests these sections through the real snapshot loader.
It also inspects the actual model request for harness guidance, prompt order, and a live global prompt edit.
extensions/subagents/api.ts
+2 -2
View File
@@ -38,7 +38,7 @@ export const Subagents = defineDoc<{
}
>
}>({
kind: 'ultimate.subagents',
kind: 'standing.subagents',
checkpointWhen: () => false,
version: 1,
scope: 'conversation',
@@ -74,7 +74,7 @@ export const Runs = defineDoc<Record<string, Run>>({
* made on the fly has none: `adhoc`, so nothing looks a definition up by its label.
*/
export const Profile = defineDoc<{ definition?: string; adhoc?: true }>({
kind: 'ultimate.agent-profile',
kind: 'standing.agent-profile',
checkpointWhen: () => false,
version: 1,
scope: 'conversation',
extensions/subagents/children.ts
+1
View File
@@ -26,6 +26,7 @@ export async function configureChild(host: any, tx: any, id: number, options: an
await configure(tx, id, childConfiguration(options))
if (options.definition) await host.need(ModelsApi).configure(tx, id, {
agent:options.definition.name,
standing:options.standing ?? false,
inheritedTools:options.parentAgent.tools.map((tool: any)=>tool.name),
excludeTools:options.excludeTools,
model:options.model,
extensions/subagents/landing.ts
+2 -2
View File
@@ -1,6 +1,6 @@
/**
* Where a page opens when no conversation is chosen (a first load, a new chapter, a remembered one that is gone):
* @ultimate's conversation, once the shared state and the standing agents are known; 'create' when @ultimate has no
* the standing agent's conversation, once the shared state and the standing agents are known; 'create' when the standing agent has no
* conversation yet, as on a fresh preview. The root is only its history.
*/
export const landing = (
@@ -9,5 +9,5 @@ export const landing = (
standing: Readonly<Record<string, number>> | null | undefined,
): number | 'create' | undefined => {
if (!loaded || chosen !== undefined || standing === undefined) return undefined
return standing?.ultimate ?? 'create'
return standing?.['$standing'] ?? 'create'
}
extensions/subagents/refs.ts
+1 -1
View File
@@ -3,4 +3,4 @@ import { defineEvent } from '@earendil-works/sausage-events'
export const Started = defineEvent<{id:number; agent:string; model?:string; thinking?:string; parent?:number; kind:'child'|'standing'|'branch'}>('subagents.started')
export type Run = {model?:string; adhoc?:true}
export const RunsDoc = {kind:'subagents.runs'}
export const StandingDoc = {kind:'ultimate.standing-agents'}
export const StandingDoc = {kind:'standing.standing-agents'}
extensions/subagents/server.ts
+14 -7
View File
@@ -15,7 +15,7 @@ const onCommit = (harness: any, callback: any) => {
return () => observation.detach()
}
const Anchor = defineTask<null, { phase: 'done' }, null>({
name: 'ultimate.subagent-anchor',
name: 'standing.subagent-anchor',
version: 1,
initial: () => ({ phase: 'done' }),
phases: {
@@ -63,7 +63,7 @@ function createReporter() {
{ phase: 'deliver' } | { phase: 'report'; report?: string; answer?: EntryId },
null
>({
name: 'ultimate.subagent-reporter',
name: 'standing.subagent-reporter',
version: 1,
initial: (input) =>
input.answer === undefined
@@ -314,15 +314,22 @@ export default function subagents(host: any) {
const parentAgent = await root.agent(ctx)
return host.harness().commit(async (tx: any) => {
const index = await tx.doc(Standing, ROOT_CONVERSATION_ID)
const existing = index[name] && await tx.conversation(index[name])
const primary = name === found.standing
const key = primary ? '$standing' : name
const existing = index[key] && await tx.conversation(index[key])
const lifecycle = host.want({name:'conversations'})
if (existing) {
await lifecycle?.setStanding(tx, existing.id, true)
await lifecycle?.setAutomaticSummaries(tx, existing.id, false)
await configureChild(host, tx, existing.id, {definition, parentAgent})
if (primary) for (const alias of Object.keys(index)) if (alias !== '$standing' && index[alias] === existing.id) delete index[alias]
index[name] = existing.id
const metadata = await tx.doc(host.metadata, existing.id)
if (metadata.label !== name) Object.assign(metadata, {label:name, title:`@${name}`})
await configureChild(host, tx, existing.id, {definition, parentAgent, standing:primary})
return existing.id
}
const child = await createChild(host, tx, {owner: {kind:'ownerless'}, definition, parentAgent, metadata:host.metadata, label:name, title:`@${name}`})
const child = await createChild(host, tx, {owner: {kind:'ownerless'}, definition, parentAgent, metadata:host.metadata, label:name, title:`@${name}`, standing:primary})
index[key] = child.id
index[name] = child.id
await lifecycle?.setStanding(tx, child.id, true)
await lifecycle?.setAutomaticSummaries(tx, child.id, false)
@@ -428,10 +435,10 @@ export default function subagents(host: any) {
const found = await models.discover('/src/sausage')
if (!found.errors.length) await host.harness().commit(async (tx: any) => {
const index = await tx.doc(Standing, ROOT_CONVERSATION_ID)
for (const name of Object.keys(index)) if (!found.definitions.some((d: any) => d.name === name && d.resident)) {
for (const name of Object.keys(index)) if (name !== '$standing' && !found.definitions.some((d: any) => d.name === name && d.resident)) {
const id = index[name]
delete index[name]
if (await tx.conversation(id)) {
if (id !== index['$standing'] && await tx.conversation(id)) {
delete (await tx.doc(host.metadata, id)).label
await host.want({name:'conversations'})?.setStanding(tx, id, false)
}
extensions/todos/README.md
+4 -4
View File
@@ -1,15 +1,15 @@
# Todos
The root document `todos` stores the world's to-do list.
Each item belongs to `user` or `ultimate`.
Each item belongs to `user` or `standing`.
The `todo` tool adds items, completes items, and lists open items by owner.
Text must contain 1–300 characters after the tool trims it.
The list retains the newest 50 completed items. The exported `mark` helper also reopens items.
Only the standing `ultimate` conversation uses the tool.
The reader uses the existing `ultimate.standing-agents` document contract.
Only the standing `standing` conversation uses the tool.
The reader uses the existing `standing.standing-agents` document contract.
Until subagents supplies that document, no conversation qualifies automatically.
This port does not designate the root as `ultimate`.
This port does not designate the root as `standing`.
The `beforeInput` hook places a note only when the standing conversation's open list changes.
The note and the remembered list commit with the input.
extensions/todos/server.ts
+12 -12
View File
@@ -1,4 +1,4 @@
/** The world's root to-do list, its todo tool, and changed-list notes for the standing @ultimate conversation. */
/** The world's root to-do list, its todo tool, and changed-list notes for the standing conversation. */
import { int, obj, oneOf, opt, str } from '@earendil-works/pi-schema'
import {
defineDoc,
@@ -11,10 +11,10 @@ import {
import type { ConversationId, Tx } from '@earendil-works/pi-durable'
// Read subagents' standing-agent contract without requiring its extension yet.
export const Standing = defineDoc<Record<string, ConversationId>>({
kind: 'ultimate.standing-agents', checkpointWhen: () => false, version: 1,
kind: 'standing.standing-agents', checkpointWhen: () => false, version: 1,
scope: 'conversation', history: 'latest', fork: 'initial', initial: () => ({}),
})
export type Todo = { id: number; text: string; owner: 'user' | 'ultimate'; at: number; done?: number }
export type Todo = { id: number; text: string; owner: 'user' | 'standing'; at: number; done?: number }
/** Done to-dos kept; older ones are dropped when another is ticked. */
export const DONE_LIMIT = 50
@@ -29,7 +29,7 @@ export const Todos = defineDoc<{ next: number; items: Todo[] }>({
fork: 'initial',
initial: () => ({ next: 1, items: [] }),
})
/** The open list @ultimate last heard, so it hears again only after a change. */
/** The open list the standing agent last heard, so it hears again only after a change. */
const Seen = defineDoc<{ key?: string }>({
kind: 'todos.seen',
checkpointWhen: () => false,
@@ -41,14 +41,14 @@ const Seen = defineDoc<{ key?: string }>({
})
export const docs = [Todos]
const OWNERS = ['user', 'ultimate'] as const
const OWNERS = ['user', 'standing'] as const
export async function add(tx: Tx, text: unknown, owner: unknown) {
if (typeof text !== 'string' || !text.trim()) throw new Error('text must be nonempty')
if (text.trim().length > TEXT_CHARS)
throw new Error(`text must be at most ${TEXT_CHARS} characters`)
if (!OWNERS.includes(owner as Todo['owner']))
throw new Error("owner must be 'user' or 'ultimate'")
throw new Error("owner must be 'user' or 'standing'")
const todos = await tx.doc(Todos, ROOT_CONVERSATION_ID)
const todo: Todo = {
id: todos.next++,
@@ -85,7 +85,7 @@ const line = (todo: Todo) => `- ${todo.id}: ${todo.text}`
export function listText(items: readonly Todo[]) {
const open = items.filter((item) => item.done === undefined)
if (!open.length) return undefined
const mine = open.filter((item) => item.owner === 'ultimate')
const mine = open.filter((item) => item.owner === 'standing')
const theirs = open.filter((item) => item.owner === 'user')
return [
...(mine.length ? ['Yours:', ...mine.map(line)] : []),
@@ -93,7 +93,7 @@ export function listText(items: readonly Todo[]) {
].join('\n')
}
/** @ultimate's note when the open list changed since it last heard it; undefined otherwise. */
/** the standing agent's note when the open list changed since it last heard it; undefined otherwise. */
export async function noteFor(tx: Tx, conversationId: ConversationId, now: number) {
const items = (await tx.doc(Todos, ROOT_CONVERSATION_ID)).items
const text = listText(items)
@@ -121,7 +121,7 @@ export async function noteFor(tx: Tx, conversationId: ConversationId, now: numbe
}
const standing = (tx: Tx, id: ConversationId) =>
tx.doc(Standing, ROOT_CONVERSATION_ID).then((doc) => doc.ultimate === id)
tx.doc(Standing, ROOT_CONVERSATION_ID).then((doc) => doc['$standing'] === id)
export default function todosExtension() {
return defineExtension({
@@ -141,11 +141,11 @@ export default function todosExtension() {
name: 'todo',
deferred: true,
description:
"The world's to-dos, owned by you ('ultimate') or the user ('user'): add {text, owner}, done {id}, list. Keep open loops here. Only the standing @ultimate conversation maintains this list.",
"The world's to-dos, owned by you ('standing') or the user ('user'): add {text, owner}, done {id}, list. Keep open loops here. Only the standing conversation maintains this list.",
parameters: obj({
action: oneOf(['add', 'done', 'list']),
text: opt(str()),
owner: opt(oneOf(['user', 'ultimate'])),
owner: opt(oneOf(['user', 'standing'])),
id: opt(int({ minimum: 1 })),
}),
// Not replay-safe (the default): an interrupted add is never rerun, so it commits at most once.
@@ -156,7 +156,7 @@ export default function todosExtension() {
// Deferred only hides it: a branch can still find it with tool_search, so the caller is checked here.
if (!(await standing(tx, caller)))
throw new Error(
'Only @ultimate keeps the to-dos; ask @ultimate to add one.',
'Only the standing agent keeps the to-dos; ask the standing agent to add one.',
)
if (args.action === 'add') {
const todo = await add(tx, args.text, args.owner)
extensions/watchdog/README.md
+3 -3
View File
@@ -6,13 +6,13 @@ A temporary provider failure creates a durable `watchdog.failure` task. The task
A failed continuation does not receive another retry. Permanent errors report immediately.
If another run remains active at the deadline, watchdog does not submit a continuation.
Reports use the messages service with a stable request ID. Owned children report to their owner. Other conversations report to @ultimate.
A failure of @ultimate itself produces a notice instead. The service retains the latest 20 reported failures.
Reports use the messages service with a stable request ID. Owned children report to their owner. Other conversations report to the standing agent.
A failure of the standing agent itself produces a notice instead. The service retains the latest 20 reported failures.
Receipts and a high-water mark prevent repeated recovery. Fresh installation excludes historical failures.
On reopen, watchdog recovers failures that its observer missed. Retry inputs also use stable request IDs.
Before @ultimate receives an input, watchdog can add an idle-branch note. It does not modify the system prompt.
Before the standing agent receives an input, watchdog can add an idle-branch note. It does not modify the system prompt.
The note excludes active branches, concluded branches, user questions, and queued landings when workspaces is available.
It waits 30 minutes, or two hours while a branch awaits an agent reply. It repeats only when the branch IDs change.
extensions/watchdog/idle.ts
+1 -1
View File
@@ -58,7 +58,7 @@ export async function idleNote(
return false
}
const idle = []
// Branches are ownerless, so subagents are skipped before any document read: this runs before every @ultimate input.
// Branches are ownerless, so subagents are skipped before any document read: this runs before every the standing agent input.
for (const conversation of conversations) {
if (conversation.owner || excluded.has(conversation.id)) continue
if (!(await tx.doc(Branch, conversation.id)).branch) continue
extensions/watchdog/retry.ts
+1 -1
View File
@@ -128,7 +128,7 @@ import type { AssistantMessage } from '@earendil-works/pi-ai'
/**
* Whether a model error is a rate limit Pi retries: HTTP 429, "rate limit", "too many requests". Retrying sooner only
* hits the limit again, so Ultimate pauses until the user presses Play, and the held run continues through Pi's retry.
* hits the limit again, so Standing agent pauses until the user presses Play, and the held run continues through Pi's retry.
* Quota, usage-limit and billing errors are on Pi's no-retry list: they end the run at once and fail as usual.
* Overloaded servers (529) are the provider's capacity, not the account's limit.
*/
extensions/watchdog/server.ts
+4 -4
View File
@@ -310,10 +310,10 @@ export default function watchdog(host: ExtensionHost) {
)
return
const child = (await tx.conversation(task.conversationId))!
const ultimate = (await tx.doc(Standing, ROOT_CONVERSATION_ID)).ultimate
const recipient = child.owner?.conversationId ?? ultimate
const standing = (await tx.doc(Standing, ROOT_CONVERSATION_ID))['$standing']
const recipient = child.owner?.conversationId ?? standing
if (recipient === undefined)
throw new Error('Watchdog needs @ultimate to report a failed run')
throw new Error('Watchdog needs the standing agent to report a failed run')
if (task.state.status !== 'terminal' || task.state.outcome.status !== 'failed') return
const error = task.state.outcome.error.message
const { hint, temporary } = diagnose(error, info.assistant)
@@ -412,7 +412,7 @@ export default function watchdog(host: ExtensionHost) {
hooks: [
hook(GenerationTask, {
beforeInput: async (_input, { tx, conversationId }) =>
(await tx.doc(Standing, ROOT_CONVERSATION_ID)).ultimate === conversationId
(await tx.doc(Standing, ROOT_CONVERSATION_ID))['$standing'] === conversationId
? idleNote(tx, host, conversationId, Date.now())
: undefined,
}),
extensions/workspaces/api.ts
+1 -1
View File
@@ -4,7 +4,7 @@ import { defineEvent } from '@earendil-works/sausage-events'
export type Workspace = {source:string;target:string;path:string;branch:string;base?:string;state:'preparing'|'ready'|'failed'|'removed';preparation?:number;error?:string;landed?:string}
export type Submission = {id:string;conversationId:number;source:string;target:string;tip:string;at:number;divergence:string;title:string;conversations:string[];prepared?:{base:string;sha:string};landed?:{base:string;sha:string;commits:{sha:string;subject:string}[];paths:string[]};outcome?:any}
export const Workspaces = defineDoc<{items:Record<string,Workspace>;queue:Submission[];landers:Record<string,number>;landerConversations:Record<string,number>;results:Record<string,any>}>({
kind:'ultimate.workspaces',version:1,scope:'conversation',history:'latest',fork:'initial',checkpointWhen:()=>false,
kind:'standing.workspaces',version:1,scope:'conversation',history:'latest',fork:'initial',checkpointWhen:()=>false,
initial:()=>({items:{},queue:[],landers:{},landerConversations:{},results:{}}),
})
export const WorkspacesApi = {name:'workspaces'}
host/js/README.md
+2 -2
View File
@@ -37,7 +37,7 @@ CARGO_TARGET_DIR=target/harness nix develop -c cargo test -p sausage --test harn
The lock file pins npm dependencies. The bundle command requires Nushell, GNU patch, and npm.
It applies `patches/pi-durable-1.5.0.patch` before esbuild runs.
This is Ultimate's patch at `7ec0b4f`, without the Node execution environment hunks (D65).
This is Standing agent's patch at `7ec0b4f`, without the Node execution environment hunks (D65).
The patch adds transactional `beforeInput`, input markers, scheduler hold/release, and the failed-run boundary fix.
The step verifies an existing application before another bundle build. A mismatched or partially applied patch fails the command.
`npm ci` restores the unpatched dependency. The next bundle command applies the patch again.
@@ -85,5 +85,5 @@ The signal client sends `Input::Stop`. The harness closes its watches, credentia
The host then stops its process services and HTTP listener.
Console output still streams assistant messages, without replay of the historical transcript.
`harness_foundation.rs` runs the three Ultimate input tests and the scheduler phase-boundary test through bundled V8 modules.
`harness_foundation.rs` runs the three Standing agent input tests and the scheduler phase-boundary test through bundled V8 modules.
It also tests the channel with closed stdin, CLI EOF and SIGTERM, submission scans, and the patched bundle hooks.
host/js/automations-pause-cases.ts
+9 -9
View File
@@ -1,4 +1,4 @@
/** All 19 Ultimate pause cases, executed against world snapshot modules. */
/** All 19 Standing agent pause cases, executed against world snapshot modules. */
import {expect,it,until,setImmediate} from "./automations-test-support";
import {fixture} from "./automations-pause-fixture";
import {BACKGROUND_CONTEXT as context} from "@earendil-works/chord/context";
@@ -245,9 +245,9 @@ it('pauses on the first rate-limited answer: the run waits in its retry, fails n
},
])
const { root, harness } = await f.open()
const ultimate = await harness.createConversation({ ownership: { kind: 'ownerless' } }, context)
const standing = await harness.createConversation({ ownership: { kind: 'ownerless' } }, context)
await harness.commit(async (tx) => {
;(await tx.doc(Standing, ROOT_CONVERSATION_ID)).ultimate = ultimate.id
;(await tx.doc(Standing, ROOT_CONVERSATION_ID))['$standing'] = standing.id
}, context)
const submission = await root.submit({ type: 'input', content: 'go' }, context)
await until(async () => (await f.action('state')).limit !== undefined)
@@ -263,7 +263,7 @@ it('pauses on the first rate-limited answer: the run waits in its retry, fails n
expect(generation.record.state.status).not.toBe('terminal')
expect((await harness.snapshot(LiveDoc, root.id, context))?.run).toBeDefined()
const cards = async () =>
(await ultimate.context(context)).entries.filter((entry) => entry.kind === 'pause.limit')
(await standing.context(context)).entries.filter((entry) => entry.kind === 'pause.limit')
await until(async () => (await cards()).length === 1)
expect((await cards())[0].data).toMatchObject({ error: limit, conversationId: root.id })
@@ -420,9 +420,9 @@ it('plays a rate-limit pause by itself once, N minutes in, only when the user sw
},
])
const { root, harness } = await f.open()
const ultimate = await harness.createConversation({ ownership: { kind: 'ownerless' } }, context)
const standing = await harness.createConversation({ ownership: { kind: 'ownerless' } }, context)
await harness.commit(async (tx) => {
;(await tx.doc(Standing, ROOT_CONVERSATION_ID)).ultimate = ultimate.id
;(await tx.doc(Standing, ROOT_CONVERSATION_ID))['$standing'] = standing.id
}, context)
const submission = await root.submit({ type: 'input', content: 'go' }, context)
await until(async () => (await f.action('state')).limit !== undefined)
@@ -443,14 +443,14 @@ it('plays a rate-limit pause by itself once, N minutes in, only when the user sw
expect(responses).toBe(2)
// Once: the pause is over, so a second due timer finds nothing to play.
expect(await f.autoPlay(since! + 31 * 60_000)).toBe(false)
const played = (await ultimate.context(context)).entries.filter(
const played = (await standing.context(context)).entries.filter(
(entry) => entry.kind === 'pause.autoplay',
)
expect(played).toHaveLength(1)
expect(played[0].data).toMatchObject({ since, at: since! + 30 * 60_000, minutes: 30 })
const notes = (await root.context(context)).entries.filter((entry) => entry.kind === 'pause.note')
expect(notes[0].data).toMatchObject({ auto: true })
expect(notes[0].model?.[0]?.content).toMatch(/continued after Ultimate played by itself\.$/)
expect(notes[0].model?.[0]?.content).toMatch(/continued after Standing agent played by itself\.$/)
// The setting stays for the next limit.
expect((await f.action('state')).autoPlay).toBe(30)
})
@@ -495,7 +495,7 @@ const at = (kind: string, phase: string) => ({ kind, state: { checkpoint: { phas
it('treats message deliveries and routes as passive waiters', () => {
expect(passive(at('board.delivery', 'deliver'))).toBe(true)
expect(passive(at('board.delivery', 'reply'))).toBe(true)
expect(passive(at('ultimate.subagent-reporter', 'deliver'))).toBe(true)
expect(passive(at('standing.subagent-reporter', 'deliver'))).toBe(true)
expect(passive(at('board.route', 'observe'))).toBe(true)
expect(passive(at('conversations.summary', 'generate'))).toBe(false)
})
host/js/automations-pause-fixture.ts
+1 -1
View File
@@ -1,4 +1,4 @@
/** Pause cases use snapshot modules and isolated durable storage, as Ultimate's fixture does. */
/** Pause cases use snapshot modules and isolated durable storage, as Standing agent's fixture does. */
import { BACKGROUND_CONTEXT as context } from '@earendil-works/chord/context'
import { obj } from '@earendil-works/pi-schema'
import { createModels } from '@earendil-works/pi-ai/models'
host/js/automations-schedule-cases.ts
+1 -1
View File
@@ -1,4 +1,4 @@
/** The five Ultimate calendar cases execute inside deno_core's V8. */
/** The five Standing agent calendar cases execute inside deno_core's V8. */
import {describe, expect, it} from "./automations-test-support";
const {describeSchedule,nextAt,parseSchedule}=await import(`${globalThis.automationBase}/automations/schedule.ts`);
const time = Date.parse('2026-01-05T08:00:00Z')
host/js/automations-test-bundle.js
+7579 -367
View File
File diff suppressed because it is too large Load Diff
host/js/automations-test-support.ts
+1 -1
View File
@@ -1,4 +1,4 @@
/** Minimal assertions for the imported Ultimate cases, without Node or Vitest. */
/** Minimal assertions for the imported Standing agent cases, without Node or Vitest. */
export const cases: {name:string;run:()=>any}[]=[]
export const it=(name:string,run:()=>any)=>cases.push({name,run})
export const describe=(_name:string,run:()=>any)=>run()
host/js/bundle.js
+7702 -1061
View File
File diff suppressed because it is too large Load Diff
host/js/chain-test-bundle.js
+2467 -200
View File
File diff suppressed because it is too large Load Diff
host/js/core-ext-test-bundle.js
+7037 -364
View File
File diff suppressed because it is too large Load Diff
host/js/core-ext-test.ts
+28 -9
View File
@@ -5,6 +5,7 @@ import { fauxProvider, fauxAssistantMessage } from '@earendil-works/pi-ai/provid
import { Harness, AgentDoc, createRegistry } from '@earendil-works/pi-durable'
import { openStorage } from './sqlite'
import { loadExtensions, fs } from './extensions'
import { parse } from 'yaml'
const assert = (ok:any,message:string)=>{if(!ok)throw new Error(message);Deno.core.ops.op_output(`ok - ${message}\n`)}
const models=createModels(), faux=fauxProvider({tokensPerSecond:100000});models.setProvider(faux.provider)
const proxyTemplate={...models.getModels('faux')[0],provider:'openai',baseUrl:'https://proxy.invalid/v1'}
@@ -18,17 +19,35 @@ const loader=await loadExtensions(harness,registry,storage,models,context)
try {
assert(!loader.state().error,`loader: ${loader.state().error ?? 'loaded'}`)
const {selection,conversations,State,metadata}=globalThis.core
const config = parse(await fs.read('/etc/sausage/config.yaml'))
config.standing='worker'
config.roster=[{name:'worker',description:'Worker',model:'faux/faux-1',thinking:'low',tools:['read','write'],prompt:'Worker body'}]
const save = ()=>fs.write('/etc/sausage/config.yaml',JSON.stringify(config))
await save()
const worker=await selection.start({agent:'worker',cwd:'/project/sub'},context)
let agent=await worker.agent(context)
assert(agent.model.modelId==='faux-1' && agent.thinkingLevel==='low','worker fast family and thinking')
assert(agent.tools.map(t=>t.name).join(',')==='read,write','worker tools and project overlay')
assert(agent.instructions.includes('Worker body') && agent.instructions.includes('Project body'),'definition body and overlay')
assert(agent.model.modelId==='faux-1' && agent.thinkingLevel==='low','worker roster model and thinking')
assert(agent.tools.map(t=>t.name).join(',')==='read,write','worker roster tools')
assert(agent.instructions.indexOf('## Sausage and the world') < agent.instructions.indexOf('Worker body'),'global prompt precedes role prompt')
const sections:any={}
for(const section of agent.sections) sections[section.key]=await section.render({conversationId:worker.id,agent},context)
assert(sections.project_context.includes('Root instructions') && sections.project_context.includes('Nested instructions'),'cwd and parent AGENTS.md')
assert(!sections.project_context.includes('Outside instructions'),'instructions stop at repo root')
assert(sections.skills.includes('builtin-skill') && sections.skills.includes('project-skill'),'builtin and project skills')
assert(sections.cwd==='/project/sub' && sections.tools.includes('read'),'cwd and tools sections')
const checkPrompt = (ctx: any, expected: string) => {
const prompt=JSON.stringify(ctx.messages.filter((message: any)=>message.role==='system'))
for (const route of ['whoami','conversations/ID/entries?after=N','events?after=N','models','extensions','credential-requests','/proc/processes','/proc/extensions','jaq']) assert(prompt.includes(route),`fresh system prompt includes ${route}`)
assert(prompt.includes(expected),'system prompt uses current global prompt')
assert(prompt.indexOf(expected)<prompt.indexOf('Worker body'),'system prompt places global prompt before role prompt')
}
faux.setResponses([(ctx)=>{checkPrompt(ctx,'## Sausage and the world');return fauxAssistantMessage('fresh')}])
harness.resume()
await (await worker.submit({type:'input',content:'fresh'},context)).wait(context)
config.prompt='Edited global prompt\\n'+config.prompt
await save()
faux.setResponses([(ctx)=>{checkPrompt(ctx,'Edited global prompt');return fauxAssistantMessage('edited')}])
await (await worker.submit({type:'input',content:'edited'},context)).wait(context)
const seed=await harness.commit(tx=>tx.appendEntry(worker.id,{kind:'test.seed'}),context)
const child=await worker.fork(seed.id,{ownership:{kind:'ownerless'}},context)
const unrelated=await harness.createConversation({ownership:{kind:'ownerless'}},context)
@@ -46,7 +65,7 @@ try {
const receipt=await child.submit({type:'input',content:'resume'},context)
assert(JSON.stringify([...await conversations.concluded(context)])===JSON.stringify([unrelated.id]),'input placement atomically resumes ancestors')
harness.resume();await receipt.wait(context)
await fs.write('/src/sausage/agents/models.json',JSON.stringify({default:{model:'faux/faux-1'},fast:{model:'faux/faux-1',thinking:'high'}}))
config.roster[0].thinking='high';await save()
faux.setResponses([fauxAssistantMessage('next')])
await (await worker.submit({type:'input',content:'next'},context)).wait(context)
agent=await worker.agent(context)
@@ -54,22 +73,22 @@ try {
// A second faux model proves selection changes, not only the thinking level.
const provider=models.getProvider('faux'), template=models.getModels('faux')[0]
models.setProvider({...provider,getModels:()=>[template,{...template,id:'second'}]})
await fs.write('/src/sausage/agents/models.json',JSON.stringify({default:{model:'faux/faux-1'},fast:{model:'faux/second'}}))
config.roster[0].model='faux/second';await save()
faux.setResponses([fauxAssistantMessage('second')])
await (await worker.submit({type:'input',content:'second'},context)).wait(context)
assert((await harness.snapshot(AgentDoc,worker.id,context)).model.modelId==='second','edited family changes next model without restart')
await fs.write('/src/sausage/agents/models.json',JSON.stringify({default:{model:'faux/faux-1'},fast:{model:'openai/new'}}))
assert((await harness.snapshot(AgentDoc,worker.id,context)).model.modelId==='second','edited roster changes next model without restart')
config.roster[0].model='openai/new';await save()
faux.setResponses([fauxAssistantMessage('proxy')])
await (await worker.submit({type:'input',content:'proxy'},context)).wait(context)
assert(models.getModel('openai','new').baseUrl===proxyTemplate.baseUrl && models.getModel('openai','new').cost.input===0 && models.getProvider('openai').auth===proxyAuth,'new model retains proxy transport and auth')
await fs.write('/src/sausage/agents/models.json',JSON.stringify({default:{model:'faux/faux-1'},fast:{model:'anthropic/new'}}))
config.roster[0].model='anthropic/new';await save()
faux.setResponses([fauxAssistantMessage('anthropic')])
await (await worker.submit({type:'input',content:'anthropic'},context)).wait(context)
assert(models.getModel('anthropic','new').baseUrl==='https://proxy.invalid/apis/anthropic','new Anthropic model retains its own provider transport')
globalThis.modelPin={provider:'faux',modelId:'faux-1'}
faux.setResponses([fauxAssistantMessage('pin')])
await (await worker.submit({type:'input',content:'pin'},context)).wait(context)
assert((await worker.agent(context)).model.provider==='faux','models.pin overrides family selection')
assert((await worker.agent(context)).model.provider==='faux','models.pin overrides roster selection')
await harness.commit(tx=>conversations.setStanding(tx,worker.id,true),context)
let refused=false;try{await harness.commit(tx=>conversations.concludeIn(tx,[worker.id]),context)}catch{refused=true}
assert(refused,'standing conversation cannot conclude')
host/js/dream-test-bundle.js
+7001 -356
View File
File diff suppressed because it is too large Load Diff
host/js/dream-test.ts
+1 -1
View File
@@ -35,7 +35,7 @@ const {Automations}=await import(`${base}/automations/server.ts`)
const {ContextUsageDoc}=await import(`${base}/automations/triggers.ts`)
const {Fired}=await import(`${base}/automations/refs.ts`)
const {Messages}=await import(`${base}/messages/api.ts`)
const primary=(await harness.conversation(await globalThis.dreamSubagents.standing('ultimate',context),context))!
const primary=(await harness.conversation(await globalThis.dreamSubagents.standing('standing',context),context))!
const entries=async()=> (await primary.context(context)).entries
const ledger=async()=> (await harness.snapshot(Automations,context))?.automations??{}
const requested=async()=> (await entries()).filter((e:any)=>e.kind==='dream.requested')
host/js/events-test-bundle.js
+7000 -355
View File
File diff suppressed because it is too large Load Diff
host/js/events.md
+1 -1
View File
@@ -42,7 +42,7 @@ Event emission updates the feed inside the same transaction.
Notices become `sausage.notice` events.
The host records each distinct notice once per facet load.
This task does not add Ultimate's build or supervisor startup events.
This task does not add Standing agent's build or supervisor startup events.
## Tests
host/js/extensions-test-bundle.js
+7000 -355
View File
File diff suppressed because it is too large Load Diff
host/js/extensions.ts
+8 -2
View File
@@ -1,5 +1,6 @@
// World extension snapshots, staged services, lifecycle, and last-good recovery.
import * as durable from '@earendil-works/pi-durable';
import * as yaml from 'yaml';
import * as tools from '@earendil-works/pi-durable/tools';
import * as ai from '@earendil-works/pi-ai';
import * as modelApi from '@earendil-works/pi-ai/models';
@@ -68,7 +69,7 @@ export async function git(cwd:string,...args:string[]) {
return result.stdout.trim();
}
export function registerModules() {
const modules = {'@earendil-works/sausage-events':events,'@earendil-works/pi-schema':schema,'@earendil-works/pi-durable':durable,'@earendil-works/pi-durable/tools':tools,'@earendil-works/pi-ai':ai,'@earendil-works/pi-ai/models':modelApi,'@earendil-works/chord':chord,'@earendil-works/chord/context':contextApi};
const modules = {'yaml':yaml,'@earendil-works/sausage-events':events,'@earendil-works/pi-schema':schema,'@earendil-works/pi-durable':durable,'@earendil-works/pi-durable/tools':tools,'@earendil-works/pi-ai':ai,'@earendil-works/pi-ai/models':modelApi,'@earendil-works/chord':chord,'@earendil-works/chord/context':contextApi};
globalThis.__sausage = {modules};
ops.op_modules(Object.fromEntries(Object.entries(modules).map(([name,value])=>[name,Object.keys(value)])));
}
@@ -209,7 +210,12 @@ export async function loadExtensions(harness:any, registry:any, storage:any, mod
const rid = ops.op_watch([`${repo}/.git/refs/heads/main`,`${repo}/.git/packed-refs`,`${repo}/settings.json`]);
await enqueue();
const pending = (async()=> {while(active){await ops.op_watch_next(rid);if(active)await enqueue();}})();
return {reload:enqueue, state:()=>({...state}), async invoke(name:string,key:string,args:any,signal:AbortSignal,conversationId?:number) {
return {reload:enqueue, state:()=>({...state}), async standing() {
if (!providers.has('models') || !providers.has('subagents')) return undefined;
const {standing:name}=await providers.get('models').discover(repo);
const conversationId=await providers.get('subagents').standing(name,context);
return {name,conversationId};
}, async invoke(name:string,key:string,args:any,signal:AbortSignal,conversationId?:number) {
const facet = facets.find(f=>f.extension.name===name);
const action = facet?.actions?.[key];
if (!action) throw Object.assign(new Error('Unknown user action'),{status:404});
host/js/foundation-test-bundle.js
+4 -3
View File
@@ -20614,7 +20614,7 @@ async function openStorage(path2) {
}
// input.ts
async function serve(harness2, context, afterSubmit, api) {
async function serve(harness2, context, afterSubmit, api, defaultConversation) {
const pending = /* @__PURE__ */ new Set();
const track2 = (work) => {
const task = work.catch((error) => Deno.core.ops.op_output(`${String(error)}
@@ -20637,8 +20637,9 @@ async function serve(harness2, context, afterSubmit, api) {
track2(api.handle(input));
continue;
}
const conversation = await harness2.conversation(input.conversation, context);
if (!conversation) throw new Error(`Unknown conversation: ${input.conversation}`);
const id = defaultConversation ? await defaultConversation() : input.conversation;
const conversation = await harness2.conversation(id, context);
if (!conversation) throw new Error(`Unknown conversation: ${id}`);
const submission2 = await conversation.submit({
type: "input",
content: input.content,
host/js/foundation-test.ts
+1 -1
View File
@@ -1,4 +1,4 @@
// Ultimate's input-placement cases, scheduler hold, and the headless channel contract.
// Standing agent's input-placement cases, scheduler hold, and the headless channel contract.
import { BACKGROUND_CONTEXT as context } from '@earendil-works/chord/context';
import { createModels } from '@earendil-works/pi-ai/models';
import { fauxProvider, fauxAssistantMessage, fauxToolCall } from '@earendil-works/pi-ai/providers/faux';
host/js/goal-cases.ts
+35 -35
View File
@@ -1,4 +1,4 @@
/** The seven Ultimate goal cases use overlaid documents and the loaded server modules. */
/** The seven Standing agent goal cases use overlaid documents and the loaded server modules. */
import type { ConversationId, GenerationHooks, Tx } from '@earendil-works/pi-durable'
import { BACKGROUND_CONTEXT } from '@earendil-works/chord/context'
import { expect, it } from './goal-questions-test-support'
@@ -6,7 +6,7 @@ const { reminderTask } = await import(`${globalThis.goalBase}/goal/reminder.ts`)
const {default:goalExtension,act,actions,end,Goals,openGoal,start,TRIES} = await import(`${globalThis.goalBase}/goal/server.ts`)
import type { Host } from './extensions.ts'
const ULTIMATE = 5 as ConversationId
const STANDING = 5 as ConversationId
const screenshot = { type: 'image', data: 'aGk=', mimeType: 'image/png' }
/** Like the harness's document overlay, a proxy at every level: structuredClone refuses it. */
@@ -18,17 +18,17 @@ const overlay = <T extends object>(value: T): T =>
},
})
/** A transaction over overlaid documents and a few entries of @ultimate's conversation; @ultimate is conversation 5. */
/** A transaction over overlaid documents and a few entries of @standing's conversation; @standing is conversation 5. */
function fakeTx(
newer: { id: number; conversationId: number; kind?: string; model: object[] }[] = [],
) {
const docs = new Map<string, object>()
const entries = [
...newer,
{ id: 30, conversationId: ULTIMATE, model: [{ role: 'user', content: 'no images' }] },
{ id: 30, conversationId: STANDING, model: [{ role: 'user', content: 'no images' }] },
{
id: 20,
conversationId: ULTIMATE,
conversationId: STANDING,
model: [{ role: 'toolResult', content: [{ type: 'text', text: 'shot' }, screenshot] }],
},
{ id: 10, conversationId: 9, model: [{ role: 'user', content: [screenshot] }] },
@@ -47,7 +47,7 @@ function fakeTx(
if (!docs.has(key))
docs.set(
key,
def.kind === 'ultimate.standing-agents' ? { ultimate: ULTIMATE } : def.initial(),
def.kind === 'standing.standing-agents' ? { '$standing': STANDING } : def.initial(),
)
return overlay(docs.get(key)!)
},
@@ -120,27 +120,27 @@ it('keeps one goal open at a time and ends it once', async () => {
it('blocks a criterion after repeated failed tries and finishes with the outcome', async () => {
const tx = fakeTx()
await expect(act(tx, ULTIMATE, { action: 'met', n: 1 }, 0)).rejects.toThrow('No hands-off goal')
await expect(act(tx, STANDING, { action: 'met', n: 1 }, 0)).rejects.toThrow('No hands-off goal')
await start(tx, 'Goal', 0)
await act(tx, ULTIMATE, { action: 'criteria', criteria: ['Loads in 1s', 'Tests pass'] }, 0)
await expect(act(tx, ULTIMATE, { action: 'met', n: 3 }, 0)).rejects.toThrow('No criterion 3')
for (let i = 1; i < TRIES; i++) await act(tx, ULTIMATE, { action: 'failed', n: 1 }, 0)
await act(tx, STANDING, { action: 'criteria', criteria: ['Loads in 1s', 'Tests pass'] }, 0)
await expect(act(tx, STANDING, { action: 'met', n: 3 }, 0)).rejects.toThrow('No criterion 3')
for (let i = 1; i < TRIES; i++) await act(tx, STANDING, { action: 'failed', n: 1 }, 0)
expect((await openGoal(tx))!.criteria[0]).toMatchObject({ status: 'open', tries: TRIES - 1 })
const last = await act(tx, ULTIMATE, { action: 'failed', n: 1 }, 0)
const last = await act(tx, STANDING, { action: 'failed', n: 1 }, 0)
expect(last.answer).toContain('blocked')
expect((await openGoal(tx))!.criteria[0].status).toBe('blocked')
// Rewording keeps what was learned about an unchanged criterion.
await act(tx, ULTIMATE, { action: 'criteria', criteria: ['Loads in 1s', 'Lint passes'] }, 0)
await act(tx, STANDING, { action: 'criteria', criteria: ['Loads in 1s', 'Lint passes'] }, 0)
expect((await openGoal(tx))!.criteria.map((item) => item.status)).toEqual(['blocked', 'open'])
await expect(act(tx, ULTIMATE, { action: 'finish', summary: 'x' }, 0)).rejects.toThrow('outcome')
await expect(act(tx, STANDING, { action: 'finish', summary: 'x' }, 0)).rejects.toThrow('outcome')
const done = await act(
tx,
ULTIMATE,
STANDING,
{ action: 'finish', outcome: 'blocked', summary: 'Stuck' },
4,
)
expect(done.ended).toMatchObject({
ended: { by: 'ultimate', outcome: 'blocked', summary: 'Stuck' },
ended: { by: 'standing', outcome: 'blocked', summary: 'Stuck' },
})
expect(await openGoal(tx)).toBeUndefined()
})
@@ -149,11 +149,11 @@ it('records decisions with the screenshots of its own conversation only', async
const tx = fakeTx()
await start(tx, 'Goal', 0)
await expect(
act(tx, ULTIMATE, { action: 'decide', question: 'Go?', picked: 'Yes' }, 0),
act(tx, STANDING, { action: 'decide', question: 'Go?', picked: 'Yes' }, 0),
).rejects.toThrow('reason')
await act(
tx,
ULTIMATE,
STANDING,
{
action: 'decide',
question: 'Approve?',
@@ -165,7 +165,7 @@ it('records decisions with the screenshots of its own conversation only', async
1,
)
await expect(
act(tx, ULTIMATE, { action: 'decide', question: 'q', picked: 'p', reason: 'r', images: 10 }, 2),
act(tx, STANDING, { action: 'decide', question: 'q', picked: 'p', reason: 'r', images: 10 }, 2),
).rejects.toThrow('has no images')
expect((await openGoal(tx))!.decisions).toEqual([
{ at: 1, question: 'Approve?', picked: 'Yes', reason: 'Meets 1', conversation: 7, images: 20 },
@@ -173,17 +173,17 @@ it('records decisions with the screenshots of its own conversation only', async
})
it('never takes an earlier turn\'s screenshot for images: "latest"', async () => {
const tx = fakeTx([{ id: 40, kind: 'pi.user', conversationId: ULTIMATE, model: [] }])
const tx = fakeTx([{ id: 40, kind: 'pi.user', conversationId: STANDING, model: [] }])
await start(tx, 'Goal', 0)
const decide = { action: 'decide', question: 'q', picked: 'p', reason: 'r' } as const
await expect(act(tx, ULTIMATE, { ...decide, images: 'latest' }, 1)).rejects.toThrow(
await expect(act(tx, STANDING, { ...decide, images: 'latest' }, 1)).rejects.toThrow(
'no images in this turn',
)
await act(tx, ULTIMATE, { ...decide, images: 20 }, 2)
await act(tx, STANDING, { ...decide, images: 20 }, 2)
expect((await openGoal(tx))!.decisions).toMatchObject([{ images: 20 }])
})
it("notes the open goal on @ultimate's runs only, and lets only @ultimate keep it", async () => {
it("notes the open goal on @standing's runs only, and lets only @standing keep it", async () => {
const tx = fakeTx()
const extension = goalExtension({
doc: () => undefined,
@@ -193,10 +193,10 @@ it("notes the open goal on @ultimate's runs only, and lets only @ultimate keep i
const hooks = extension.hooks![0].handlers as GenerationHooks
const before = (conversationId: number, kind = 'pi.user') =>
hooks.beforeInput({ kind }, { tx, conversationId } as never, BACKGROUND_CONTEXT)
expect(await before(ULTIMATE)).toBeUndefined()
expect(await before(STANDING)).toBeUndefined()
await start(tx, 'Make it fast', 0)
expect(await before(9)).toBeUndefined()
const note = (await before(ULTIMATE))!.entries![0]
const note = (await before(STANDING))!.entries![0]
expect(note.kind).toBe('goal.note')
expect(note.model![0].content).toContain(
'Hands-off goal 1, granted by the user, who is away: Make it fast',
@@ -209,19 +209,19 @@ it("notes the open goal on @ultimate's runs only, and lets only @ultimate keep i
{ conversationId, commit: (fn: (tx: Tx) => unknown) => fn(tx) } as never,
undefined as never,
)
await expect(run(9, { action: 'met', n: 1 })).rejects.toThrow('Only @ultimate')
expect(await run(ULTIMATE, { action: 'criteria', criteria: ['Fast'] })).toEqual({
await expect(run(9, { action: 'met', n: 1 })).rejects.toThrow('Only the standing agent')
expect(await run(STANDING, { action: 'criteria', criteria: ['Fast'] })).toEqual({
content: [{ type: 'text', text: 'Goal 1 has 1 criteria.' }],
})
expect((await tx.doc(Goals, 1 as ConversationId)).goals[0].criteria).toHaveLength(1)
// Finishing ends the goal and creates the report's delivery in the same commit; the note stops.
await run(ULTIMATE, { action: 'finish', outcome: 'met', summary: 'Done' })
await run(STANDING, { action: 'finish', outcome: 'met', summary: 'Done' })
expect(tx.tasks).toEqual([
{ name: 'goal.deliver', input: expect.objectContaining({ to: ULTIMATE, kind: 'goal.report' }) },
{ name: 'goal.deliver', input: expect.objectContaining({ to: STANDING, kind: 'goal.report' }) },
{ name: 'goal.reminder', input: { goal: 1 } },
])
expect(await before(ULTIMATE)).toBeUndefined()
expect(await before(STANDING)).toBeUndefined()
})
it('keeps a reminder while the goal is open: paused while waiting on the user, gone once ended', async () => {
@@ -232,7 +232,7 @@ it('keeps a reminder while the goal is open: paused while waiting on the user, g
const run = (args: object) =>
tool.execute(
args as never,
{ conversationId: ULTIMATE, commit: (fn: (tx: Tx) => unknown) => fn(tx) } as never,
{ conversationId: STANDING, commit: (fn: (tx: Tx) => unknown) => fn(tx) } as never,
undefined as never,
)
await actions.start({ text: 'Make it fast' }, { host } as never)
@@ -240,7 +240,7 @@ it('keeps a reminder while the goal is open: paused while waiting on the user, g
// its key instead of making a second one, and pauses it.
await service.create({
name: 'Goal reminder · goal 1',
subscribers: ['@ultimate'],
subscribers: ['@standing'],
key: 'goal.reminder:1',
})
;(await openGoal(tx))!.waiting = { reason: 'Which account?', at: 0 }
@@ -249,7 +249,7 @@ it('keeps a reminder while the goal is open: paused while waiting on the user, g
{
id: 'A1',
name: 'Goal reminder · goal 1',
subscribers: ['@ultimate'],
subscribers: ['@standing'],
key: 'goal.reminder:1',
enabled: false,
},
@@ -261,10 +261,10 @@ it('keeps a reminder while the goal is open: paused while waiting on the user, g
// The user renamed it: it stays linked by its id.
ledger.get('A1')!.name = 'Night check'
// Its fire tells @ultimate what to check; other inputs do not.
// Its fire tells @standing what to check; other inputs do not.
const hooks = extension.hooks![0].handlers as GenerationHooks
const note = async (input: object) =>
(await hooks.beforeInput(input as never, { tx, conversationId: ULTIMATE }, BACKGROUND_CONTEXT))!
(await hooks.beforeInput(input as never, { tx, conversationId: STANDING }, BACKGROUND_CONTEXT))!
.entries![0].model![0].content as string
expect(await note({ kind: 'automations.fired', data: { id: 'A1' } })).toContain(
'This is the goal reminder',
@@ -315,7 +315,7 @@ it('grants nothing through a start card still queued after the user ended its go
expect(
await hooks.beforeInput(
{ kind: 'goal.started' },
{ tx, conversationId: ULTIMATE },
{ tx, conversationId: STANDING },
BACKGROUND_CONTEXT,
),
).toBeUndefined()
host/js/goal-questions-fixture.ts
+2 -2
View File
@@ -30,8 +30,8 @@ export async function fixture(clock: any = {}) {
const settled=async()=>{await until(async()=>(await harness.commit((tx:any)=>tx.scanTasks({},1000),context)).items.every((t:any)=>t.state.status==='terminal'));await harness.waitForIdle(context)}
const entries=async(id:number)=>(await harness.commit((tx:any)=>tx.scanEntries({conversationId:id},1000),context)).items.toSorted((a:any,b:any)=>a.id-b.id)
const type=async(id:number,content:string)=>{const c=await harness.conversation(id,context);await(await c.submit({type:'input',content},context)).wait(context)}
const ultimate=async()=>{const c=await harness.createConversation({ownership:{kind:'ownerless'}},context);await c.configure({model:{provider:'faux',modelId:'faux-1'}},context);await harness.commit(async(tx:any)=>{(await tx.doc(Standing,root.id)).ultimate=c.id},context);return c.id}
const standing=async()=>{const c=await harness.createConversation({ownership:{kind:'ownerless'}},context);await c.configure({model:{provider:'faux',modelId:'faux-1'}},context);await harness.commit(async(tx:any)=>{(await tx.doc(Standing,root.id))['$standing']=c.id},context);return c.id}
const ask=async(id:number,args:any)=>{await type(id,`SEND ${JSON.stringify({to:'user',...args})}`);await settled();const result=(await entries(id)).findLast((e:any)=>e.kind==='pi.tool-result')?.model?.[0];return {error:result?.isError?result.content.map((p:any)=>p.text).join(''):undefined,message:(await log()).messages.findLast((m:any)=>m.from===id&&m.to==='user')}}
const action=(conversationId?:number)=>({host:globalThis.testHost,conversationId,signal:new AbortController().signal})
return {base,faux,loader,host:globalThis.testHost,opened:{harness,root,close:async()=>{await loader.close();await harness.close(context);await storage.close()}},ultimate,type,ask,answer:(args:any)=>question.actions.answer(args,action()),dismiss:(messageId:number)=>question.actions.dismiss({messageId},action()),log,entries,observed,settled,hear:(fn:any)=>heard=fn}
return {base,faux,loader,host:globalThis.testHost,opened:{harness,root,close:async()=>{await loader.close();await harness.close(context);await storage.close()}},standing,type,ask,answer:(args:any)=>question.actions.answer(args,action()),dismiss:(messageId:number)=>question.actions.dismiss({messageId},action()),log,entries,observed,settled,hear:(fn:any)=>heard=fn}
}
host/js/goal-questions-test-bundle.js
+7613 -401
View File
File diff suppressed because it is too large Load Diff
host/js/goal-questions-test-support.ts
+1 -1
View File
@@ -1,4 +1,4 @@
/** Minimal assertions for the imported Ultimate cases, without Node or Vitest. */
/** Minimal assertions for the imported Standing agent cases, without Node or Vitest. */
export const cases: {name:string;run:()=>any}[]=[]
export const it=(name:string,run:()=>any)=>cases.push({name,run})
export const describe=(_name:string,run:()=>any)=>run()
host/js/goal-questions-test.ts
+1 -1
View File
@@ -1,4 +1,4 @@
/** Ultimate goal and question-card cases execute in the world loader. */
/** Standing agent goal and question-card cases execute in the world loader. */
import {fixture} from './goal-questions-fixture'
import {cases,cleanups} from './goal-questions-test-support'
const initial=await fixture()
host/js/goal-reminder-cases.ts
+1 -1
View File
@@ -9,7 +9,7 @@ it('fires one goal reminder through automations',async()=>{
const timer={at:now+delay,callback};timers.add(timer);return()=>timers.delete(timer)
}})
try {
const id=await f.ultimate();const {harness}=f.opened
const id=await f.standing();const {harness}=f.opened
const goal=await import(/* @vite-ignore */ `${f.base}/goal/server.ts`)
const {Automations}=await import(/* @vite-ignore */ `${f.base}/automations/server.ts`)
await goal.actions.start({text:'Ship the goal'},{host:f.host})
host/js/harness-api-test-bundle.js
+7012 -374
View File
File diff suppressed because it is too large Load Diff
host/js/harness-api-test.ts
+3 -2
View File
@@ -14,11 +14,12 @@ const curl=async(path:string,body?:any)=>{
assert(result.code===0,`curl failed: ${result.stderr}`);
return JSON.parse(result.stdout);
};
const id=await f.ultimate();
const id=await f.standing();
const question=await f.ask(id,{text:'Choose.',questions:[{text:'Name',kind:'text',ref:'1'}]});
assert(question.message,'Question fixture');
const listed=await curl('/conversations');
assert(listed.conversations.some((c:any)=>c.id===id && c.agent && !c.running),'Conversation list');
assert(listed.standing.name==='standing'&&listed.standing.conversationId===id,'Conversation list identifies the configured standing conversation');
const initial=await curl(`/conversations/${id}/entries`);
const after=initial.entries.at(-1).id;
const answer=await curl('/extensions/questions/actions/answer',{args:{messageId:question.message.id,answers:[{question:'Name',answer:'API choice'}]}});
@@ -26,7 +27,7 @@ assert(!answer.error,JSON.stringify(answer));
await f.settled();
assert(JSON.stringify(await f.entries(id)).includes('API choice'),'Question answer reaches asker');
assert((await curl('/models')).models.length>0,'Models');
assert((await curl('/families')).families.default,'Families');
assert((await curl('/families')).error,'The obsolete families route is absent');
assert((await curl('/extensions')).good,'Extension status');
assert(Array.isArray((await curl('/credential-requests')).requests),'Credential requests');
assert((await curl('/unknown')).error,'Unknown route');
host/js/harness-api.ts
+3 -6
View File
@@ -1,6 +1,6 @@
// The world chat API runs in the same harness as turns and extension actions.
import { AgentDoc, LiveDoc, defineDoc } from '@earendil-works/pi-durable';
import { Metadata, fs } from './extensions';
import { Metadata } from './extensions';
import { eventLog } from './events';
const Concluded = defineDoc({kind:'conversations.concluded',version:1,scope:'session',checkpointWhen:()=>false,initial:()=>({ids:[]})});
const Started = defineDoc({kind:'board.started',version:1,scope:'conversation',history:'latest',fork:'initial',checkpointWhen:()=>false,initial:()=>({})});
@@ -51,6 +51,7 @@ export function harnessApi(harness:any, models:any, extensions:any, context:any)
const reply=(status:number,body:any)=>{send(input.id,{status,body});send(input.id,null);};
try {
if(input.method==='GET' && path==='/conversations') {
const standing=await extensions.standing?.() ?? null;
const records=await harness.commit(async(tx:any)=>{
const all:any[]=[];let cursor:any;
do {const page=await tx.scanConversations({},256,cursor);all.push(...page.items);cursor=page.next;}while(cursor);
@@ -64,17 +65,13 @@ export function harnessApi(harness:any, models:any, extensions:any, context:any)
const started:any=await harness.snapshot(Started,record.id,context);
return {...record,title:metadata?.title??'',agent,starter:started?.by??null,concluded:concluded.has(record.id),running:!!live?.run};
}));
reply(200,{conversations});return;
reply(200,{conversations,standing});return;
}
if(input.method==='GET' && path==='/events') {
const cursor=integer(url.searchParams.get('after'),0);
await follow(input,async(after)=>log.recent({since:after,order:'ascending',limit:256}),cursor);return;
}
if(input.method==='GET' && path==='/extensions') {reply(200,extensions.state());return;}
if(input.method==='GET' && path==='/families') {
let families={};try{families=JSON.parse(await fs.read('/src/sausage/agents/models.json'));}catch(error){if((error as any).code!=='not_found')throw error;}
reply(200,{families});return;
}
if(input.method==='GET' && path==='/models') {reply(200,{models:models.getAllModels().map(({id,name,provider,contextWindow,maxTokens,reasoning,cost}:any)=>({id,name,provider,contextWindow,maxTokens,reasoning,cost}))});return;}
const action=/^\/extensions\/([^/]+)\/actions\/([^/]+)$/.exec(path);
if(input.method==='POST' && action) {
host/js/input.ts
+4 -3
View File
@@ -1,5 +1,5 @@
// EOF and Stop drain submissions. Shutdown cancels invocations and preserves durable tasks for restart.
export async function serve(harness: any, context: any, afterSubmit?: () => Promise<void>, api?: any) {
export async function serve(harness: any, context: any, afterSubmit?: () => Promise<void>, api?: any, defaultConversation?: () => Promise<number>) {
const pending = new Set<Promise<void>>();
const track = (work: Promise<void>) => {
const task = work.catch(error => Deno.core.ops.op_output(`${String(error)}\n`)).finally(() => pending.delete(task));
@@ -22,8 +22,9 @@ export async function serve(harness: any, context: any, afterSubmit?: () => Prom
track(api.handle(input));
continue;
}
const conversation = await harness.conversation(input.conversation, context);
if (!conversation) throw new Error(`Unknown conversation: ${input.conversation}`);
const id = defaultConversation ? await defaultConversation() : input.conversation;
const conversation = await harness.conversation(id, context);
if (!conversation) throw new Error(`Unknown conversation: ${id}`);
const submission = await conversation.submit({
type: 'input', content: input.content, requestId: input.requestId,
}, context);
host/js/main.ts
+13 -18
View File
@@ -1,7 +1,7 @@
// Durable root conversation and console. SQLite stays outside the world.
// The standing conversation serves the console and chat. SQLite stays outside the world.
import { BACKGROUND_CONTEXT as context } from '@earendil-works/chord/context';
import { proxyModels } from './proxy-models';
import { Harness, AgentDoc, createRegistry } from '@earendil-works/pi-durable';
import { Harness, createRegistry } from '@earendil-works/pi-durable';
import { CodingTools } from './environment';
import { openStorage } from './sqlite';
import { environment, conversationChains } from './environment';
@@ -9,10 +9,8 @@ import { CredentialRequests, startRequests } from './credentials';
import { serve } from './input';
import { harnessApi } from './harness-api';
import { loadExtensions } from './extensions';
const proxy = Deno.core.ops.op_model_proxy();
if (!proxy?.base_url || !proxy?.api_key) throw new Error('The llm-proxy credential is required');
const models = await proxyModels(proxy);
const model = models.getModel('openai', 'gpt-6.1-sol');
const proxy = await Deno.core.ops.op_model_proxy();
const models = await proxyModels(proxy, () => Deno.core.ops.op_model_proxy());
const registry = createRegistry();
registry.install(CodingTools);
registry.install(CredentialRequests);
@@ -22,14 +20,13 @@ const harness = await Harness.open(storage, {
env: async (target: any) => environment(target.cwd ?? '/', await chain(target.conversationId)),
}, context);
const chain = conversationChains(harness, context);
const root = await harness.root(context, { agent: { model: { provider: 'openai', modelId: model.id }, thinkingLevel: 'low', cwd: '/' } });
// T32 stored this exact built-in list in the console root. Use registry defaults now.
const storedAgent = await harness.snapshot(AgentDoc, root.id, context);
if (Array.isArray(storedAgent?.extensions) && storedAgent.extensions.length === 2 &&
[CodingTools, CredentialRequests].every(extension => storedAgent.extensions.includes(extension.name))) {
await root.configure({ extensions: null }, context);
}
const watch = await root.watch(context);
await harness.root(context, {agent:{cwd:'/src/sausage'}});
const stopRequests = await startRequests(harness, context);
const extensions = await loadExtensions(harness, registry, storage, models, context, [CodingTools, CredentialRequests]);
const standing = await extensions.standing();
if (!standing) throw new Error('The world needs the models and subagents extensions for its standing agent');
const conversation = (await harness.conversation(standing.conversationId, context))!;
const watch = await conversation.watch(context);
const emitted = new Map<number, string>();
let partial = '';
const text = (message: any) => message?.content?.filter((b: any) => b.type === 'text').map((b: any) => b.text).join('') ?? '';
@@ -52,17 +49,15 @@ for (const entry of watch.value.entries) {
if (entry.kind === 'pi.assistant') emitted.set(entry.id, text(entry.model?.[0]));
}
if (globalThis.consoleEnabled) watch.start((event: any) => { if (event.kind === 'value') render(event.value); });
const stopRequests = await startRequests(harness, context);
const extensions = await loadExtensions(harness, registry, storage, models, context, [CodingTools, CredentialRequests]);
const api = harnessApi(harness, models, extensions, context);
harness.resume();
await serve(harness, context, globalThis.consoleEnabled ? async () => {
await harness.waitForIdle(context);
const final = await root.watch(context);
const final = await conversation.watch(context);
render(final.value);
await final.stop();
Deno.core.ops.op_output('\n');
} : undefined, api);
} : undefined, api, async () => (await extensions.standing())!.conversationId);
await api.close();
await extensions.close();
await stopRequests();
host/js/messages-test-bundle.js
+7001 -356
View File
File diff suppressed because it is too large Load Diff
host/js/messages-test.ts
+1 -1
View File
@@ -59,7 +59,7 @@ const {SubmissionLanded} = await import(`${base}/lineage/refs.ts`);
const {Tally, abstract} = await import(`${base}/lineage/server.ts`);
const {normalizeMemoryOp, parseMemoryOps, applyMemoryOps, readMemory, memorizeTask} = await import(`${base}/lineage/memory.ts`);
const {Seen, mainOf} = await import(`${base}/lineage/catchup.ts`);
const primaryId = await globalThis.subagentService.standing('ultimate', context);
const primaryId = await globalThis.subagentService.standing('standing', context);
const primary = (await harness.conversation(primaryId, context))!;
const entries = async (conversation: any) => (await conversation.context(context)).entries;
const replies = async () => (await entries(primary)).filter((e: any) => e.kind === 'board.reply');
host/js/p3-test-bundle.js
+7012 -367
View File
File diff suppressed because it is too large Load Diff
host/js/p3-test.ts
+12 -12
View File
@@ -12,10 +12,10 @@ testReplay();
if (globalThis.p3Mode !== 'unit') {
const workerId = (transcript: any) => Number(/Started subagent (\d+)/.exec(JSON.stringify(transcript.messages))?.[1]);
const rules: Rule[] = [
{ id: 'ultimate-branch', role: 'ultimate', match: { kind: 'user', text: 'launch scenario' }, answer: { tools: [{ name: 'branch', args: { repo: '/src/demo', isolated: true, prompt: 'begin' } }] } },
{ id: 'ultimate-probe', role: 'ultimate', match: { kind: 'user', text: 'launch branch' }, answer: { tools: [{ name: 'branch', args: { repo: '/src/demo', prompt: 'branch probe' } }] } },
{ id: 'standing-branch', role: 'standing', match: { kind: 'user', text: 'launch scenario' }, answer: { tools: [{ name: 'branch', args: { repo: '/src/demo', isolated: true, prompt: 'begin' } }] } },
{ id: 'standing-probe', role: 'standing', match: { kind: 'user', text: 'launch branch' }, answer: { tools: [{ name: 'branch', args: { repo: '/src/demo', prompt: 'branch probe' } }] } },
{ id: 'coordinator-probe', role: 'coordinator', match: { kind: 'user', text: 'branch probe' }, answer: { text: 'branch complete' } },
{ id: 'ultimate-receipt', role: 'ultimate', match: { text: '' }, answer: { text: 'Receipt noted.' } },
{ id: 'standing-receipt', role: 'standing', match: { text: '' }, answer: { text: 'Receipt noted.' } },
{ id: 'coordinator-echo', role: 'coordinator', match: { kind: 'user', text: 'Receipt noted.' }, answer: { text: 'Acknowledged.' } },
{ id: 'start-worker', role: 'coordinator', match: { kind: 'user', text: 'begin' }, answer: { tools: [{ name: 'subagent', args: { action: 'start', agent: 'worker', isolate: 'worktree', task: 'build feature', background: true } }] } },
{ id: 'start-reviewer', role: 'coordinator', match: { kind: 'user', text: 'worker complete' }, answer: { tools: [{ name: 'subagent', args: { action: 'start', agent: 'reviewer', isolate: 'none', task: 'review feature', background: true } }] } },
@@ -32,7 +32,7 @@ if (globalThis.p3Mode !== 'unit') {
models.setProvider(scriptedModel(rules, rule => Deno.core.ops.op_output(`scripted rule: ${rule}\n`)));
const storage = await openStorage(globalThis.storagePath), registry = createRegistry();
const harness = await Harness.open(storage, { models, registry, settings: { retry: { enabled: false }, get extensions() { return registry.snapshot().installed(); } } }, context);
const root = await harness.root(context, { agent: { model: { provider: 'faux', modelId: 'faux-1' }, instructions: 'P3 role:ultimate', cwd: '/src/demo' } });
const root = await harness.root(context, { agent: { model: { provider: 'faux', modelId: 'faux-1' }, instructions: 'P3 role:standing', cwd: '/src/demo' } });
// Stop after the durable lander commit records landed.sha, before its event/report commit.
if (globalThis.p3Mode === 'kill-forward') {
const install = registry.install.bind(registry);
@@ -85,12 +85,12 @@ if (globalThis.p3Mode !== 'unit') {
if (globalThis.p3Mode === 'load-all') {
assert((await root.agent(context)).tools.some((tool: any) => tool.name === 'branch'), 'Every extension must load');
} else {
const ultimate = (await harness.conversation(await globalThis.p3Subagents.standing('ultimate', context), context))!;
const standing = (await harness.conversation(await globalThis.p3Subagents.standing('standing', context), context))!;
harness.resume();
const probe = globalThis.p3Mode.startsWith('branch');
if (globalThis.p3Mode === 'branch' || globalThis.p3Mode.startsWith('kill-')) {
await (await ultimate.submit({ type: 'input', content: probe ? 'launch branch' : 'launch scenario', requestId: 'p3-launch' }, context)).wait(context);
const entries = (await ultimate.context(context)).entries;
await (await standing.submit({ type: 'input', content: probe ? 'launch branch' : 'launch scenario', requestId: 'p3-launch' }, context)).wait(context);
const entries = (await standing.context(context)).entries;
assert(!JSON.stringify(entries).includes('"isError":true'), `Branch tool failed: ${JSON.stringify(entries)}`);
}
if (globalThis.p3Mode.startsWith('kill-')) await new Promise(() => {});
@@ -101,12 +101,12 @@ if (globalThis.p3Mode !== 'unit') {
await new Promise(resolve => setTimeout(resolve, 0));
}
};
const replies = async () => (await ultimate.context(context)).entries.filter((entry: any) => entry.kind === 'board.reply');
const replies = async () => (await standing.context(context)).entries.filter((entry: any) => entry.kind === 'board.reply');
await wait(async () => (await replies()).length > 0, 'No branch reply');
const received = await replies();
const coordinator = (await harness.conversation(received[0].data.from, context))!;
assert((await coordinator.agent(context)).instructions === 'P3 role:coordinator', 'Branch must use coordinator definition');
assert((await conversationChains(harness, context)(coordinator.id)).includes(String(ultimate.id)), 'D63 must include ultimate');
assert((await coordinator.agent(context)).instructions.includes('P3 role:coordinator'), 'Branch must use coordinator definition');
assert((await conversationChains(harness, context)(coordinator.id)).includes(String(standing.id)), 'D63 must include standing');
if (!probe) {
await wait(async () => Object.keys((await harness.snapshot(globalThis.p3Workspaces.doc, root.id, context)).results).length === 1, 'No landing');
await wait(async () => (await harness.snapshot(globalThis.p3Workspaces.doc, root.id, context)).queue.length === 0, 'Lander did not report');
@@ -125,9 +125,9 @@ if (globalThis.p3Mode !== 'unit') {
assert(!await fs.stat(worker.path).then(() => true, () => false), 'Released checkout must not exist');
assert(Object.values(index.results).every((result: any) => result.outcome === 'landed'), `Landing failed: ${JSON.stringify(index)}`);
assert((await globalThis.p3Host.eventLog.recent({ events: [{ kind: 'event.workspaces.landed' }] })).length === 1, 'Expected one landed event');
const ultimateEntries = (await ultimate.context(context)).entries;
const standingEntries = (await standing.context(context)).entries;
const landed = Object.values(index.results)[0] as any;
const cards = ultimateEntries.filter((entry: any) => entry.kind === 'board.message' && entry.data.text.startsWith('[lander] Submission') && entry.data.text.includes(' landed as '));
const cards = standingEntries.filter((entry: any) => entry.kind === 'board.message' && entry.data.text.startsWith('[lander] Submission') && entry.data.text.includes(' landed as '));
assert(cards.length === 1 && cards[0].data.from === coordinator.id && cards[0].data.text.includes(landed.sha), `Expected one landing card through messages: ${JSON.stringify(cards)}`);
const repliesNow = await replies();
assert(repliesNow.length === 1 && !repliesNow[0].data.failed && new Set(repliesNow.map((entry: any) => entry.data.id)).size === repliesNow.length, `Duplicate board.reply: ${JSON.stringify(repliesNow)}`);
host/js/p4-test-bundle.js
+1 -1
View File
@@ -1697,7 +1697,7 @@ trap - EXIT
printf 'P4_PHASE_2_OK
'`,
`set -e
undo list --conversation 1 > /tmp/p4-undo.json
undo list > /tmp/p4-undo.json
box ls > /tmp/p4-boxes
printf 'P4_PHASE_3_OK
'`
host/js/p4-test.ts
+1 -1
View File
@@ -56,7 +56,7 @@ box rm "$id"
trap - EXIT
printf 'P4_PHASE_2_OK\n'`,
`set -e
undo list --conversation 1 > /tmp/p4-undo.json
undo list > /tmp/p4-undo.json
box ls > /tmp/p4-boxes
printf 'P4_PHASE_3_OK\n'`,
];
host/js/package-lock.json
Generated
+17 -1
View File
@@ -7,7 +7,8 @@
"dependencies": {
"@earendil-works/chord": "1.5.0",
"@earendil-works/pi-ai": "1.5.0",
"@earendil-works/pi-durable": "1.5.0"
"@earendil-works/pi-durable": "1.5.0",
"yaml": "2.9.1"
},
"devDependencies": {
"esbuild": "0.28.2"
@@ -1682,6 +1683,21 @@
"optional": true
}
}
},
"node_modules/yaml": {
"version": "2.9.1",
"resolved": "https://registry.npmjs.org/yaml/-/yaml-2.9.1.tgz",
"integrity": "sha512-3NxN8+78OdzbT7C/WjGsyfPAtJaN3FNDsWxv7Y7mcDsT/oOmgW8BpyQQFFBnvZE3j9Y2Sdz1ULFLezL7Eb2yFw==",
"license": "ISC",
"bin": {
"yaml": "bin.mjs"
},
"engines": {
"node": ">= 14.6"
},
"funding": {
"url": "https://github.com/sponsors/eemeli"
}
}
}
}
host/js/package.json
+11 -6
View File
@@ -1,11 +1,16 @@
{
"private": true,
"type": "module",
"scripts": { "bundle": "nu apply-patch.nu && esbuild main.ts --bundle --format=esm --platform=browser --outfile=bundle.js && esbuild test.ts --bundle --format=esm --platform=browser --outfile=test-bundle.js && esbuild durable-test.ts --bundle --format=esm --platform=browser --outfile=durable-test-bundle.js && esbuild foundation-test.ts --bundle --format=esm --platform=browser --outfile=foundation-test-bundle.js && esbuild extensions-test.ts --bundle --format=esm --platform=browser --outfile=extensions-test-bundle.js && esbuild events-test.ts --bundle --format=esm --platform=browser --outfile=events-test-bundle.js && esbuild subagents-test.ts --bundle --format=esm --platform=browser --outfile=subagents-test-bundle.js && esbuild core-ext-test.ts --bundle --format=esm --platform=browser --outfile=core-ext-test-bundle.js && esbuild plan-todos-test.ts --bundle --format=esm --platform=browser --outfile=plan-todos-test-bundle.js && esbuild p2-test.ts --bundle --format=esm --platform=browser --outfile=p2-test-bundle.js && esbuild chain-test.ts --bundle --format=esm --platform=browser --outfile=chain-test-bundle.js && esbuild undo-harness-test.ts --bundle --format=esm --platform=browser --outfile=undo-harness-test-bundle.js && esbuild messages-test.ts --bundle --format=esm --platform=browser --outfile=messages-test-bundle.js && esbuild p3-test.ts --bundle --format=esm --platform=browser --outfile=p3-test-bundle.js && esbuild automations-test.ts --bundle --format=esm --platform=browser --outfile=automations-test-bundle.js && esbuild watchdog-test.ts --bundle --format=esm --platform=browser --outfile=watchdog-test-bundle.js && esbuild workspaces-test.ts --bundle --format=esm --platform=browser --outfile=workspaces-test-bundle.js && esbuild goal-questions-test.ts --bundle --format=esm --platform=browser --outfile=goal-questions-test-bundle.js && esbuild dream-test.ts --bundle --format=esm --platform=browser --outfile=dream-test-bundle.js && esbuild p4-test.ts --bundle --format=esm --platform=browser --outfile=p4-test-bundle.js && esbuild harness-api-test.ts --bundle --format=esm --platform=browser --outfile=harness-api-test-bundle.js && esbuild proxy-models-test.ts --bundle --format=esm --platform=browser --outfile=proxy-models-test-bundle.js" },
"dependencies": {
"@earendil-works/pi-durable": "1.5.0",
"@earendil-works/pi-ai": "1.5.0",
"@earendil-works/chord": "1.5.0"
"scripts": {
"bundle": "nu apply-patch.nu && esbuild main.ts --bundle --format=esm --platform=browser --outfile=bundle.js && esbuild test.ts --bundle --format=esm --platform=browser --outfile=test-bundle.js && esbuild durable-test.ts --bundle --format=esm --platform=browser --outfile=durable-test-bundle.js && esbuild foundation-test.ts --bundle --format=esm --platform=browser --outfile=foundation-test-bundle.js && esbuild extensions-test.ts --bundle --format=esm --platform=browser --outfile=extensions-test-bundle.js && esbuild events-test.ts --bundle --format=esm --platform=browser --outfile=events-test-bundle.js && esbuild subagents-test.ts --bundle --format=esm --platform=browser --outfile=subagents-test-bundle.js && esbuild core-ext-test.ts --bundle --format=esm --platform=browser --outfile=core-ext-test-bundle.js && esbuild plan-todos-test.ts --bundle --format=esm --platform=browser --outfile=plan-todos-test-bundle.js && esbuild p2-test.ts --bundle --format=esm --platform=browser --outfile=p2-test-bundle.js && esbuild chain-test.ts --bundle --format=esm --platform=browser --outfile=chain-test-bundle.js && esbuild undo-harness-test.ts --bundle --format=esm --platform=browser --outfile=undo-harness-test-bundle.js && esbuild messages-test.ts --bundle --format=esm --platform=browser --outfile=messages-test-bundle.js && esbuild p3-test.ts --bundle --format=esm --platform=browser --outfile=p3-test-bundle.js && esbuild automations-test.ts --bundle --format=esm --platform=browser --outfile=automations-test-bundle.js && esbuild watchdog-test.ts --bundle --format=esm --platform=browser --outfile=watchdog-test-bundle.js && esbuild workspaces-test.ts --bundle --format=esm --platform=browser --outfile=workspaces-test-bundle.js && esbuild goal-questions-test.ts --bundle --format=esm --platform=browser --outfile=goal-questions-test-bundle.js && esbuild dream-test.ts --bundle --format=esm --platform=browser --outfile=dream-test-bundle.js && esbuild p4-test.ts --bundle --format=esm --platform=browser --outfile=p4-test-bundle.js && esbuild harness-api-test.ts --bundle --format=esm --platform=browser --outfile=harness-api-test-bundle.js && esbuild proxy-models-test.ts --bundle --format=esm --platform=browser --outfile=proxy-models-test-bundle.js"
},
"devDependencies": { "esbuild": "0.28.2" }
"dependencies": {
"@earendil-works/chord": "1.5.0",
"@earendil-works/pi-ai": "1.5.0",
"@earendil-works/pi-durable": "1.5.0",
"yaml": "2.9.1"
},
"devDependencies": {
"esbuild": "0.28.2"
}
}

Some files were not shown because too many files have changed in this diff Show More