eeymoo/claw-code
1
0
Fork 0
mirror of https://github.com/instructkr/claw-code.git synced 2026-04-13 11:34:51 +08:00
Code Issues Packages Projects Releases Wiki Activity

Keep completed lanes from ending on mushy stop summaries

Browse Source 
The next repo-local sweep target was ROADMAP #69: completed lane
runs could persist vague control text like “commit push everyting,
keep sweeping $ralph”, which made downstream stop summaries
operationally useless. The fix adds a lane-finished quality floor
that preserves strong summaries, rewrites empty/control-only/too-
short-without-context summaries into a contextual fallback, and
records structured metadata explaining when the fallback fired.

Constraint: Keep legitimate concise lane summaries intact while improving only low-signal completions
Rejected: Blanket-rewrite every completed summary into a templated sentence | would erase useful model-authored detail from good lane outputs
Confidence: high
Scope-risk: narrow
Reversibility: clean
Directive: If lane-finished summary heuristics change later, update the structured `qualityFloorApplied/rawSummary/reasons/wordCount` contract and its regression tests together
Tested: cargo fmt --all --check; cargo clippy --workspace --all-targets -- -D warnings; cargo test --workspace; architect review APPROVE
Not-tested: External OMX consumers that may still ignore the new lane.finished data payload
This commit is contained in:
Yeachan-Heo
2026-04-12 03:22:32 +00:00
parent 17e21bc4ad
commit 26b89e583f
2 changed files with 209 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
ROADMAP.md
View File

@@ -496,7 +496,7 @@ Model name prefix now wins unconditionally over env-var presence. Regression tes
62. **Worker state file surface not implemented** — **done (verified 2026-04-12):** current `main` already wires `emit_state_file(worker)` into the worker transition path in `rust/crates/runtime/src/worker_boot.rs`, atomically writes `.claw/worker-state.json`, and exposes the documented reader surface through `claw state` / `claw state --output-format json` in `rust/crates/rusty-claude-cli/src/main.rs`. Fresh proof exists in `runtime` regression `emit_state_file_writes_worker_status_on_transition`, the end-to-end `tools` regression `recovery_loop_state_file_reflects_transitions`, and direct CLI parsing coverage for `state` / `state --output-format json`. Source: Jobdori dogfood.
**Scope note (verified 2026-04-12):** ROADMAP #31, #43, and #63-#68 currently appear to describe acpx/droid or upstream OMX/server orchestration behavior, not claw-code source already present in this repository. Repo-local searches for `acpx`, `use-droid`, `run-acpx`, `commit-wrapper`, `ultraclaw`, `roadmap-nudge-10min`, `OMX_TMUX_INJECT`, `/hooks/health`, and `/hooks/status` found no implementation hits outside `ROADMAP.md`, and the earlier state-surface note already records that the HTTP server is not owned by claw-code. With #45 now fixed, the remaining unresolved items in this section look like external tracking notes rather than confirmed repo-local backlog; re-check if new repo-local evidence appears.
**Scope note (verified 2026-04-12):** ROADMAP #31, #43, and #63-#68 currently appear to describe acpx/droid or upstream OMX/server orchestration behavior, not claw-code source already present in this repository. Repo-local searches for `acpx`, `use-droid`, `run-acpx`, `commit-wrapper`, `ultraclaw`, `roadmap-nudge-10min`, `OMX_TMUX_INJECT`, `/hooks/health`, and `/hooks/status` found no implementation hits outside `ROADMAP.md`, and the earlier state-surface note already records that the HTTP server is not owned by claw-code. With #45 and #69 now fixed, the remaining unresolved items in this section look like external tracking notes rather than confirmed repo-local backlog; re-check if new repo-local evidence appears.
63. **Droid session completion semantics broken: code arrives after "status: completed"** — dogfooded 2026-04-12. Ultraclaw droid sessions (use-droid via acpx) report `session.status: completed` before file writes are fully flushed/synced to the working tree. Discovered +410 lines of "late-arriving" droid output that appeared after I had already assessed 8 sessions as "no code produced." This creates false-negative assessments and duplicate work. **Fix shape:** (a) droid agent should only report completion after explicit file-write confirmation (fsync or existence check); (b) or, claw-code should expose a `pending_writes` status that indicates "agent responded, disk flush pending"; (c) lane orchestrators should poll for file changes for N seconds after completion before final assessment. **Blocker:** none. Source: Jobdori ultraclaw dogfood 2026-04-12.
@@ -510,6 +510,6 @@ Model name prefix now wins unconditionally over env-var presence. Regression tes
68. **Internal reinjection/resume paths leak opaque control prose** — dogfooded 2026-04-12. OMX lanes stopping with `Continue from current mode state. [OMX_TMUX_INJECT]` expose internal implementation details instead of operator-meaningful state. The event tells us *that* tmux reinjection happened, but not *why* (retry after failure? resume after idle? manual recovery?), *what state was preserved*, or *what the lane was trying to do*. **Fix shape:** recovery/reinject events should emit structured cause like: `resume_after_stop`, `retry_after_tool_failure`, `tmux_reinject_after_idle`, `manual_recovery` plus preserved state / target lane info. Never leak bare internal markers like `[OMX_TMUX_INJECT]` as the primary summary. Blocker: none. Source: gaebal-gajae dogfood analysis 2026-04-12.
69. **Lane stop summaries have no minimum quality floor** — dogfooded 2026-04-12. `clawcode-human` session stopped with summary `commit push everyting, keep sweeping $ralph` — vague, typo-ridden, operationally useless. Unlike well-scoped review lanes, this summary regressed to mushy command prose with no outcome clarity. **Fix shape:** (a) enforce minimum stop/result summary standards: what was done (outcome), what was scoped (target), what's next (state); (b) typo/grammar validation; (c) reject summaries that are shorter than N words or contain only control verbs without context. Blocker: none. Source: gaebal-gajae dogfood analysis 2026-04-12.
69. **Lane stop summaries have no minimum quality floor** — **done (verified 2026-04-12):** completed lane persistence in `rust/crates/tools/src/lib.rs` now normalizes vague/control-only stop summaries into a contextual fallback that includes the lane target and status, while preserving structured metadata about whether the quality floor fired (`qualityFloorApplied`, `rawSummary`, `reasons`, `wordCount`). Regression coverage locks both the pass-through path for good summaries and the fallback path for mushy summaries like `commit push everyting, keep sweeping $ralph`. **Original filing below.**
70. **Install-source ambiguity misleads real users** — community observation 2026-04-12. User treated `claw-code.io` as official, then hit `clawcode` vs deprecated `claw-code` naming collision and concluded install story was inconsistent. Source-of-truth is not obvious when website/repo/crates naming diverges. **Fix shape:** canonical repo docs should explicitly state which site is official; installation guidance should visibly warn against deprecated `claw-code` crate and ambiguous third-party pages. Blocker: none. Source: gaebal-gajae community watch 2026-04-12.
70. **Install-source ambiguity misleads real users** — community observation 2026-04-12. User treated `claw-code.io` as official, then hit `clawcode` vs deprecated `claw-code` naming collision and concluded install story was inconsistent. Source-of-truth is not obvious when website/repo/crates naming diverges. **Fix shape:** canonical repo docs should explicitly state which site is official; installation guidance should visibly warn against deprecated `claw-code` crate and ambiguous third-party pages. Blocker: none. Source: gaebal-gajae community watch 2026-04-12.