agentleFS
isaaclimmyisaaclimmy/codex-skill-sync/skill-sync/SKILL.md

skill-sync

Skill2 starsChanged 5 months agoMITView on GitHub

Simple sync flow for installed SKILL.md skills and Codex plugins. Use when the user says /skill-sync, check skills, update skills, run all skill updates, map skill repos, refresh Codex plugins, or create weekly skill maintenance.

21 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use session-handoff, verification-before-completion and 2 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • wpankSkill

    session-handoff

    WHAT: Create comprehensive handoff documents that enable fresh AI agents to seamlessly continue work with zero ambiguity. Solves long-running agent context exhaustion problem.

    Also in 3 projects that use this

  • OpenAISkill

    verification-before-completion

    Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

    Also in 3 projects that use this

  • OpenAISkill

    test-driven-development

    Use when implementing any feature or bugfix, before writing implementation code

    Also in 3 projects that use this

  • GoogleSkill

    tdd

    Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

    Also in 4 projects that use this

The file

Show its rules and text

We split this skill into the rules it gives an agent

27 rules. 0 are common, 22 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 17 edited versions we read keep these.

  • 22

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Keep installed skills and Codex plugins current.

    line 10
  2. Distinctive:

    If the user writes in English, respond in English.

    line 18
  3. Distinctive:

    If the user writes in Chinese, respond in Simplified Chinese.

    line 19
  4. Distinctive:

    If the user mixes English and Chinese, use the dominant language.

    line 20
  5. Distinctive:

    Keep commands, file paths, JSON keys, status values, and shell commands unchanged.

    line 21
  6. Distinctive:

    Do not show raw JSON unless the user asks.

    line 58
  7. Distinctive:

    Ask one decision at a time.

    line 60
  8. Distinctive:

    Never ask the user to edit JSON manually unless they choose advanced mode.

    line 65
  9. Distinctive:

    If editing the manifest, preserve existing entries and only add confirmed mappings.

    line 66
  10. Distinctive:

    Ask for explicit confirmation.

    line 75
  11. Distinctive:

    If Needs Setup has anything, ask whether to map repos now or skip mapping for this run.

    line 92
  12. Distinctive:

    If the user maps repos, save only confident matches.

    line 94
  13. Distinctive:

    Run a dry update check. Do not apply.

    line 95
  14. Distinctive:

    Ask for update confirmation.

    line 108
  15. Distinctive:

    After the first successful update, ask whether to create a weekly check automation.

    line 110
  16. Very specific:

    Use ~/.codex/skill-sources.json as the default saved mapping so the user does not need to map the same skills again.

    line 119
  17. Distinctive:

    Only confirmed repo mappings persist.

    line 123
  18. Distinctive:

    If the user chooses Map, public GitHub/web search is already part of the task.

    line 127
  19. Very specific:

    Read references/manifest.example.json before editing a manifest.

    line 131
  20. Distinctive:

    Search public GitHub first.

    line 135
  21. Very specific:

    Prefer the upstream public repo with strong evidence: exact SKILL.md match, same description/name, active repo, and visible community use …

    line 137
  22. Very specific:

    Do not map Needs Setup skills to catalystsystemslab/<skill>, isaaclim/<skill>, or another user/org mirror unless the installed skill itself …

    line 140
  23. Distinctive:

    If unsure, leave the repo blank or ask the user for the link.

    line 146
  24. Distinctive:

    Default to dry-run unless the user clearly asks to update.

    line 163
  25. Distinctive:

    Never overwrite Needs Setup skills.

    line 165
  26. Distinctive:

    Do not migrate a skill to a similar repo without confirmation.

    line 166
  27. Distinctive:

    Do not ask the user to override unsafe symlink or size failures.

    line 169

“Common” is measured: kept by at least 6 in 10 of the 17 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Skill Sync

Source: https://github.com/catalystsystemslab/codex-skill-sync/tree/main/skill-sync

Keep installed skills and Codex plugins current. doctor is read-only setup diagnostics. check, update, and run all all start the same inventory and mapping flow.

Language

Reply in the user's language by default.

  • If the user writes in English, respond in English.
  • If the user writes in Chinese, respond in Simplified Chinese.
  • If the user mixes English and Chinese, use the dominant language.
  • Keep commands, file paths, JSON keys, status values, and shell commands unchanged.
  • Translate explanations, summaries, warnings, and recommendations.

Important label translations:

  • Official / 官方
  • Community / 社区
  • Needs Setup / 需要设置
  • current / 已是最新
  • update_available / 有可用更新
  • updated / 已更新
  • skipped / 已跳过
  • failed / 失败

Start Here

For check, update, and run all, first list what is installed:

python3 <skill-dir>/scripts/update_codex_assets.py --inventory --json --report skill-update-report.json

Show the result as three groups. Translate internal JSON groups this way:

  • official -> Official
  • non_official -> Community
  • unmapped -> Needs Setup

Blank repo means the source is not confirmed. Do not guess.

Beginner Response Style

Most users do not know Git, JSON, branches, or subpaths.

When showing inventory:

  • Explain each group in plain language.
  • Do not show raw JSON unless the user asks.
  • Recommend the safest next action.
  • Ask one decision at a time.

When mapping:

  • Say that public GitHub search may use the skill name and description.
  • Never ask the user to edit JSON manually unless they choose advanced mode.
  • If editing the manifest, preserve existing entries and only add confirmed mappings.
  • Show exactly what will be saved before saving it.

Before apply:

  • Summarize what will change.
  • List skipped skills separately.
  • Remind the user that backups will be created.
  • Ask for explicit confirmation.

Commands

For doctor:

  1. Run read-only diagnostics.
  2. Explain warnings in plain language.
  3. Recommend the safest next action.
python3 <skill-dir>/scripts/update_codex_assets.py --doctor --report skill-doctor-report.json

For check:

  1. Run inventory.
  2. If Needs Setup has anything, ask whether to map repos now or skip mapping for this run.
  3. If the user maps repos, save only confident matches.
  4. Run a dry update check. Do not apply.
python3 <skill-dir>/scripts/update_codex_assets.py --json --report skill-update-report.json

For update or run all:

  1. Run inventory.
  2. If Needs Setup has anything, ask whether to map repos now or skip mapping for this run.
  3. If the user maps repos, save only confident matches.
  4. Show the repo list with blanks for uncertain repos.
  5. Ask for update confirmation.
  6. Apply updates.
  7. After the first successful update, ask whether to create a weekly check automation.
python3 <skill-dir>/scripts/update_codex_assets.py --apply --json --report skill-update-report.json

Mapping

Use ~/.codex/skill-sources.json as the default saved mapping so the user does not need to map the same skills again. If the user supplies another manifest, use that instead.

Only confirmed repo mappings persist. If the user skips mapping for this run, do not mark those skills local unless they explicitly say the skill is private or local-only.

If the user chooses Map, public GitHub/web search is already part of the task. Proceed with search tools and only ask again if the runtime blocks network access or a repo match is uncertain.

Read references/manifest.example.json before editing a manifest.

For each Needs Setup skill:

  • Search public GitHub first. Use the exact skill name plus unique words from its frontmatter or heading.
  • Prefer the upstream public repo with strong evidence: exact SKILL.md match, same description/name, active repo, and visible community use such as stars.
  • Do not map Needs Setup skills to catalystsystemslab/<skill>, isaaclim/<skill>, or another user/org mirror unless the installed skill itself names that repo as its source or the user confirms it.
  • Trust a repo only when it contains that skill's SKILL.md at the repo root or a clear subfolder.
  • Save confirmed matches with url, branch, and subpath.
  • If unsure, leave the repo blank or ask the user for the link.
  • Mark private/local skills as local:
{ "name": "my-private-skill", "root": "~/.codex/skills", "kind": "local" }

Status Words

  • current: already okay.
  • update_available: update found, not applied yet.
  • updated: replaced after backup.
  • skipped: unknown source: needs a repo link or local marking.
  • failed: inspect network, auth, branch, or subpath.

Safety

  • Default to dry-run unless the user clearly asks to update.
  • Back up before replacing a skill folder.
  • Never overwrite Needs Setup skills.
  • Do not migrate a skill to a similar repo without confirmation.
  • The script replaces whole skill folders; it does not merge local edits.
  • Unsafe remote skill trees must be reported as failed and never installed.
  • Do not ask the user to override unsafe symlink or size failures.
  • Codex plugin updates are Codex-only.

Automation

Offer weekly check automation only after a successful update. Weekly is the default; bi-weekly is fine for stable setups.

Weekly automation must default to dry-run check mode. Do not create an automation that runs --apply unless the user explicitly asks for automatic updates and confirms they understand it will replace mapped skill folders.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.