Commit Graph

12 Commits

Author SHA1 Message Date
Trevin Chow
9bf3b07185 fix(ce-compound): require question tool for "What's next?" prompt (#460)
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-30 16:25:26 -07:00
Trevin Chow
1f499482bc fix(git-commit-push-pr): quiet expected no-pr gh exit (#439) 2026-03-29 11:45:27 -07:00
Trevin Chow
f83305e22a fix: harden git workflow skills with better state handling (#406)
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-27 01:04:54 -07:00
Trevin Chow
b30288c44e feat: add project-standards-reviewer as always-on ce:review persona (#402) 2026-03-26 19:20:15 -07:00
Trevin Chow
7c5ff445e3 feat: promote ce:review-beta to stable ce:review (#371) 2026-03-24 21:00:38 -07:00
Trevin Chow
e932276866 feat: add ce:review-beta with structured persona pipeline (#348) 2026-03-23 21:49:04 -07:00
Trevin Chow
eaaba1928b feat: add claude-permissions-optimizer skill (#298) 2026-03-18 01:18:27 -07:00
Trevin Chow
a83e11e982 fix: review fixes — stale refs, skill counts, and validation guidance
- Fix -plan.md → -beta-plan.md in ce:plan-beta post-generation question
- Remove stale brainstorm doc reference from solutions doc
- Update plugin.json and marketplace.json skill counts (42 → 44)
- Add generic beta skill validation guidance to AGENTS.md and solutions doc
2026-03-17 10:39:02 -07:00
Trevin Chow
72d4b0dfd2 fix: add disable-model-invocation to beta skills and refine descriptions
Beta skills now use disable-model-invocation: true to prevent accidental
auto-triggering. Descriptions written as future stable descriptions with
[BETA] prefix for clean promotion. Updated solutions doc and AGENTS.md
promotion checklist to include removing the field.
2026-03-17 10:33:01 -07:00
Trevin Chow
7a81cd1aba docs: add beta skills framework pattern for parallel -beta suffix skills 2026-03-17 10:33:01 -07:00
Trevin Chow
699f484033 feat(skills): add autonomous mode to ce:compound-refresh
Support mode:autonomous argument for unattended/scheduled runs.
In autonomous mode: skip all user questions, apply safe actions
directly, mark ambiguous cases as stale with conservative confidence,
and generate a detailed report for after-the-fact human review.
2026-03-15 14:44:34 -07:00
Trevin Chow
8f4818c6e2 docs(solutions): compound learning from ce:compound-refresh skill redesign
Documents five skill design patterns discovered during testing:
platform-agnostic tool references, auto-archive consistency,
smart triage for broad scope, replacement subagents over
ce:compound handoff, and file tools over shell commands.
2026-03-15 14:44:34 -07:00