codex-skill-curator

Solid

Curate Codex skills with a Hermes-style maintenance pass: usage telemetry, stale/archive candidates, pins, backups, dry-run reports, and rollback guidance.

AI & Automation 17 stars 5 forks Updated 3 days ago MIT

Install

View on GitHub

Quality Score: 78/100

Stars 20%
42
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
92
Issue Health 10%
80
License 10%
100
Description 5%
100

Skill Content

# Codex Skill Curator Use this skill when the user asks to clean up, audit, curate, archive, restore, or review Codex skills. ## Curator Rules 1. Start with dry-run unless the user explicitly asks to mutate. 2. Only manage user-created or repo-created skills. 3. Never mutate bundled, system, admin, or plugin-distributed skills. 4. Respect pinned skills. Pinned skills can be patched, but not archived or deleted. 5. Archive instead of deleting. Archive moves must be reversible. 6. Take a backup before any mutating run. 7. Consolidate at the package level. If a skill has `references/`, `templates/`, `scripts/`, or `assets/`, preserve or re-home those files rather than flattening only `SKILL.md`. ## Tooling Prefer MCP tools from the bundled `self-improving-skills` server: - `codex_skill_curate` for stale/archive candidates (each run persists a report under the data dir's `logs/curator/` — cite it when summarizing) - `codex_skill_usage` for telemetry - `codex_skill_archive` for reversible archive moves (a name collision in `.archive/` gets a timestamp suffix automatically; restore finds it) - `codex_skill_restore` for archived skill recovery - `codex_skill_backups` / `codex_skill_rollback` to list backups and restore one by exact backup_id — the rollback backs up the current content first, so it is itself undoable (`undo_backup` in the result) - `codex_skill_prune_backups` to keep only the newest N backups per skill - `codex_skill_pin` for pin/unpin - `codex_self_imp...

Details

Author
UniM0cha
Repository
UniM0cha/self-improving-skills
Created
1 months ago
Last Updated
3 days ago
Language
Python
License
MIT

Integrates with

Similar Skills

Semantically similar based on skill content — not just same category