BERORINPO
|
db7f2a6fd5
|
fix(skills): move top-level origin frontmatter key under metadata
The official Agent Skills spec (agentskills.io/specification) whitelists exactly
6 top-level frontmatter keys (name/description/license/compatibility/metadata/
allowed-tools). A top-level `origin` key fails the official validator
(anthropics/skills quick_validate.py ALLOWED_PROPERTIES; skills-ref validate).
This moves `origin: X` -> `metadata.origin: X` across the canonical skills/
tree, preserving each value verbatim. Frontmatter-only, minimal diff.
- 251 SKILL.md updated (242 new metadata block, 9 appended to existing metadata)
- origin values preserved verbatim (verified 251/251)
- YAML validated on all changed files
- scoped to canonical skills/ only (docs/<lang> translations + tool mirrors
.cursor/.kiro/.agents left untouched; presumably regenerated from canonical)
Addresses #2233
|
2026-06-11 21:12:21 +09:00 |
|
tongshu2023
|
66ad878e68
|
feat(skills): add config-gc skill (#2216)
* feat(skills): add config-gc skill
Garbage collection for Claude Code configuration sprawl: 8 scan
channels (skills, memory, hooks, permissions, MCP, reminders,
project history, caches), confirm-each-deletion human-in-the-loop,
soft-delete with undo log. Subtractive counterpart to
workspace-surface-audit and configure-ecc.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(skills): address review feedback on config-gc
- Replace invalid comment-out strategy for JSON permission files with
backup + gc_log entry + jq array removal (cubic P1)
- Swap GNU-only find -printf for portable du -k (works on macOS/BSD)
- Capture gc date once into a variable so trash dir and undo log agree
- Simplify shadowed-permission detection with jq index() guard
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
|
2026-06-11 01:01:18 -04:00 |
|