Integration of 3 quick wins into existing UOK infrastructure:
1. Model Learning (Quick Win #2) → metrics.js
- Record outcomes to model-learner for per-task-type performance tracking
- Hook: recordUnitOutcome() now calls ModelLearner.recordOutcome()
- Fire-and-forget: never blocks outcome recording on learning failure
- Enables adaptive model routing decisions in downstream gates
2. Self-Report Fixing (Quick Win #1) → triage-self-feedback.js
- Auto-fix high-confidence reports (>0.85) in applyTriageReport()
- Hook: After triage and requirement promotion, apply auto-fixes
- Fire-and-forget: never blocks report application on fix failure
- Returns reportsAutoFixed count for triage metrics
3. Knowledge Injection (Quick Win #3) → already integrated in auto-prompts.js
- Already active in execute-task prompt template
- Semantic matching with graceful degradation
All integration points:
- Fire-and-forget: learning/fixing failures never block dispatch
- UOK-native: use existing outcome recording, db, gates
- Backward compatible: applyTriageReport now async, but callers handle it
- No new dependencies: all modules already in codebase
Testing: 2934 tests pass (no regressions from integration)
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
- sf-home.ts: new — resolves ~/.sf/ path and SF home dir helpers (port of gsd-home.ts)
- memory-embeddings.ts: new — embedding helpers for memory similarity search
- component-types.ts: new — Component, ComponentManifest, ComponentHook type defs
- workflow-install.ts: new — workflow installation from local/remote sources
- auto-post-unit.ts: clearEvidenceFromDisk after successful verification
- routing-history.ts: add cost-per-token tracking to routing decisions
- workflow-{manifest,templates}.ts: hardening sweep
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Last batch from the parallel swarm session: docstring tweaks,
verification-gate doc additions, workflow-reconcile and worktree-command
follow-ups, doctor-environment cleanup. Typecheck clean.
Most of the session work landed in earlier commits (8be8f4774, 3045538cb,
038938f2a, ed85252fc, 4f4b584e5, etc.); this commit is the residual
working-tree state after all swarms reported.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add repository-vcs-context.ts to detect and inject VCS context (Git/Jujutsu)
into the agent system prompt; wire in repo-vcs bundled skill trigger
- Add src/resources/skills/repo-vcs/ skill for commit, push, and safe-push workflows
- Add JSDoc Purpose/Consumer annotations to app-paths, bundled-extension-paths,
errors, extension-discovery, extension-registry, headless-types, headless, and traces
- Add justfile and just to flake.nix devShell
- Fill out new-user-onboarding.md spec (Draft) and core-beliefs.md (Status: Accepted)
- Add notification-event-model.md design doc and notification-source-hygiene.md spec
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
84 files spanning provider capabilities, model routing, headless
runtime, sf auto subsystems, gitbook docs, and test coverage. Snapshotted
so headless auto can resume M004 (Production Readiness) S03
(Verification Gate Validation) on a clean tree.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The single IDLE_TIMEOUT_MS constant was conflating two different jobs:
"are we done?" vs "is the agent stuck?". For multi-turn commands (auto,
next, discuss, plan), the first question is wrong — those signal
completion explicitly via "auto-mode stopped" terminal notifications,
and child-process exit catches crashes. The 120s I'd just bumped
multi-turn to was still in idle-detection mindset; that's not what we
need from this timer.
New semantics:
- IDLE_TIMEOUT_MS = 15s — quick commands (status, queue, …); idle
really does mean done.
- NEW_MILESTONE_IDLE_TIMEOUT_MS = 120s — bounded creative task with
pauses for thinking between bootstrap steps.
- MULTI_TURN_DEADLOCK_BACKSTOP_MS = 30 minutes — auto/next/discuss/plan.
Not a "done" detector; a deadlock recovery bound. Long enough to
never bother slow LLM reasoning or chained tool calls; short enough
to recover from a true hang within a reasonable window. Real
completion comes from terminal notifications + child-process exit,
both already wired.
Code reads cleaner too: effectiveIdleTimeout selection now mirrors the
three-way conceptual split.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The 15s IDLE_TIMEOUT_MS was killing auto-mode prematurely. Symptom: sf
headless auto would dispatch a task, the LLM would make 1-2 tool calls,
pause to reason about the next step, exceed 15s of "no events", and
headless would declare "Status: complete" — exiting at ~35s with the task
barely started (123 events but only 2 tool calls).
The 120s NEW_MILESTONE_IDLE_TIMEOUT_MS already exists for the same reason
("LLM may pause between tool calls e.g. after mkdir, before writing
files"). The same applies to auto/next/discuss/plan — all multi-turn
commands where the LLM thinks longer between actions, especially on
non-trivial tasks. isMultiTurnCommand was already defined for related
logic; this just wires it into the idle-timeout decision.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit captures uncommitted modifications that accumulated in the
working tree across multiple in-progress workstreams. It is a snapshot
to clear the deck before sf v3 work begins; individual workstreams
should land separately on top of this.
Notable additions:
- trace-collector.ts, traces.ts, src/tests/trace-export.test.ts —
trace export plumbing
- biome.json — Biome linter configuration
- .gitignore — exclude native/npm/**/*.node compiled binaries
The bulk of the diff is across src/resources/extensions/sf/ (301 files)
and src/resources/extensions/sf/tests/ (277 files), reflecting the
ongoing sf extension work. Specific feature commits should follow this
snapshot rather than being archaeology'd out of it.
The 76MB native/npm/linux-x64-gnu/forge_engine.node compiled binary
was left out of the commit — it's now gitignored and built locally.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Real bugs from 2nd-pass scan:
1. extension-registry.ts: discoverAllManifests skipped symlinked extension
dirs because Dirent.isDirectory() returns false for symlinks. Dev-workflow
symlinks under ~/.sf/agent/extensions/ were invisible to list/enable/
disable/info. Matches the regression documented in
symlink-extension-discovery.test.ts — the test inlines the correct logic,
but this callsite still had the buggy form. Now accepts isDirectory() ||
isSymbolicLink().
2. headless.ts SIGINT handler: client.stop() failures were double-silenced
(inner .catch(()=>{}), outer try{}catch{}). Interactive mode logs stop
errors to stderr. Restored head/headless parity — still fire-and-forget
(exit code is forced via process.exit) but failures are observable.
3. openai-codex-responses.ts SSE parser: malformed data frames were silently
dropped so broken streams looked identical to clean ones. Now debug-logs
the parse error with the chunk context so broken streams are
distinguishable in logs. Stream continues on bad chunk (one bad frame
shouldn't kill the whole generation).
4. web/cleanup-service.ts generated script: bare 'catch {}' around four native
git calls (nativeBranchList, nativeDetectMainBranch, nativeBranchListMerged,
nativeForEachRef). A failed main-branch detection silently left mainBranch
undefined-shaped, then the next native call operated on garbage. Now emits
console.warn so failures surface in the subprocess log.
5. web/undo-service.ts generated script: git revert failure was silenced;
when --no-commit failed, user saw commitsReverted=0 with no reason. Now
logs the revert error before attempting --abort (abort itself remains
best-effort silent).
False positives from the same scan (investigated and dismissed):
- auto-worktree.ts #2505: code uses ':(exclude).sf/milestones' pathspec +
shelter-and-restore, which is a better fix than the 'drop --include-untracked'
approach the test comment describes. Test comment is stale; source is correct.
- Lifecycle handler unhandled rejections across 5 extensions: extensions/runner.ts
already try/catches handler invocations and routes to emitError. Wrapping the
individual handlers would be redundant.
Three fixes to make the headless progress stream readable at a glance:
1. Filter TUI footer widget keys from setStatus — 0-emoji, 0-color-band,
authority, ollama, sf-fast, and sf-auto are sticky indicators for the
interactive TUI footer, not workflow phases. They no longer leak
through as [phase] ollama / [phase] sf-fast noise.
2. Unify tag prefix column width at 11 chars via a new tag() helper in
headless-ui.ts. All of [tool], [agent], [forge], [phase], [thinking],
[cost], [text] now align on the same column, matching the existing
[headless] and [thinking] widths.
3. Dedupe consecutive identical progress lines in headless.ts so a
widget that re-emits the same setStatus on every LLM call prints
once instead of flooding stderr. Two different lines still both show;
only adjacent duplicates collapse.
Also tightens parsePhaseLabel so an unknown bare statusKey with no
message returns null rather than leaking the raw key — a defense in
depth if the footer-widget allowlist drifts behind a new extension.
Tests: 4 new cases in headless-progress.test.ts covering footer-key
suppression, bare-key suppression, workflow-phase passthrough, and
tag-alignment. 88/88 pass.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
