prjct-cli 2.20.0 → 2.20.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +26 -0
- package/dist/bin/prjct-core.mjs +207 -207
- package/dist/daemon/entry.mjs +82 -82
- package/dist/templates.json +1 -1
- package/package.json +1 -1
- package/templates/skills/prjct/SKILL.md +7 -4
package/CHANGELOG.md
CHANGED
|
@@ -1,5 +1,31 @@
|
|
|
1
1
|
# Changelog
|
|
2
2
|
|
|
3
|
+
## [2.20.2] - 2026-05-17
|
|
4
|
+
|
|
5
|
+
### Bug Fixes
|
|
6
|
+
|
|
7
|
+
- skill routing triages complexity FIRST — spec is the exception, not the default (v2.20.1) (#341)
|
|
8
|
+
|
|
9
|
+
|
|
10
|
+
## [Unreleased]
|
|
11
|
+
|
|
12
|
+
### Changed
|
|
13
|
+
|
|
14
|
+
- **Skill routing triages complexity FIRST — spec is the exception, not the default.** `prjct-skill-body.ts` (skill SSOT) led with "substantive work → default to `spec` first", which pushed every simple one-file change through `spec` + `audit-spec` + 3 reviewer subagents — ceremony tax that slowed ship for zero protection on a fix. Inverted: an explicit **triage** step routes simple work (≈1 file, known root cause, reversible, "fix"/"hoy") DIRECT to `task` → implement → `qa`/`review` → `ship` with no spec; `spec`/`audit-spec` reserved for genuinely complex/high-stakes framing. Prose-only, no code-path change; skill-generator suite green (31/31).
|
|
15
|
+
|
|
16
|
+
## [2.20.1] - 2026-05-17
|
|
17
|
+
|
|
18
|
+
### Bug Fixes
|
|
19
|
+
|
|
20
|
+
- skill-miss-detector — crew-isolation guard (no false nags after crew runs) (v2.19.10) (#339)
|
|
21
|
+
|
|
22
|
+
|
|
23
|
+
## [Unreleased]
|
|
24
|
+
|
|
25
|
+
### Fixed
|
|
26
|
+
|
|
27
|
+
- **skill-miss-detector no longer false-positives after a crew run (#16 follow-up).** Crew implementer/reviewer run as isolated subagents in the *shared* working tree, so at the leader's Stop hook `getModifiedFiles()` saw their edits and path-overlap relevance fired — but the leader transcript never carries the memory references the subagent made in its own isolated transcript, producing a false skill-miss nag for every crew-touched file. Fix: `detectSkillMisses` collects the `files_touched` of crew runs whose `ended_at` is within `CREW_RUN_RECENCY_MS` (6h) via `crewRunStorage.list` and excludes them from path-overlap relevance; token-overlap detection stays active so non-crew work in the same session is still covered. Crew itself is unchanged (it was architecturally correct). Best-effort — any failure degrades to prior behavior. Tests: `core/__tests__/services/skill-miss-detector.test.ts`.
|
|
28
|
+
|
|
3
29
|
## [2.20.0] - 2026-05-17
|
|
4
30
|
|
|
5
31
|
### Features
|