agentleFS
busybutlazybusybutlazy/skill-forge/.agents/skills/install-manager-skill/SKILL.md

install-manager-skill

Skill1 starsChanged 7 months agoMITView on GitHub

Use this skill when a maintainer wants to install or refresh manager-only skills and shared regular skills into this repository's local Codex or Claude agent targets.

Files
SKILL.md

Install Manager Skill

用來把管理者工作流需要的 skills 同步到這個 repo 本地的 agent 目錄。

Use This For

  • 列出目前可供管理者同步的 skills
  • 從 manager-skills/ 中挑選 manager-only skills
  • 額外納入帶有 shared tag 的 regular skills
  • 讓管理者選擇要同步到 .agents/skills/、.claude/skills/ 或兩者都同步
  • 呼叫 repo CLI 完成 install、refresh 或 update

Do Not Use This For

  • 建立新的 canonical skill
  • 修改既有 canonical skill 的內容
  • 手動編輯 .agents/skills/ 或 .claude/skills/
  • 取代 create-skill 或 update-skill

Workflow

1. Inspect the available manager catalog

先整理兩類可安裝 skill:

  • canonical-skills/manager-skills/ 內全部 skills
  • canonical-skills/regular-skills/ 內帶有 shared tag 的 skills

向使用者清楚列出:

  • skill 名稱
  • 來源是 manager-skills 還是 regular-skills [shared]
  • 簡短描述
  • 若已安裝,顯示目前狀態

2. Confirm the sync target

詢問要同步到哪裡:

  • Codex:.agents/skills/
  • Claude:.claude/skills/
  • All:兩者都同步

3. Choose what to sync

讓使用者選擇:

  • 要同步哪些 skills
  • 是否要一口氣同步全部 manager catalog
  • 若有 drift 或舊的 unmanaged install,是否接受覆蓋

如果使用者沒有指定清單,預設可同步整個 manager catalog。

4. Execute through the skill-forge CLI

不要自己重寫安裝邏輯,直接呼叫 repo CLI:

  • PYTHONPATH=src python -m skill_forge --repo-root . sync-manager-catalog ...

必要時加入:

  • --project .
  • --target codex|claude|all
  • --force
  • 指定的 skill 名單

5. Report the outcome

同步完成後要回報:

  • 哪些 skills 已同步
  • 同步到哪些 target
  • 是否有任何 skip、repair、force overwrite
  • 若某 skill 不在 manager catalog,直接明講原因

Rules

  • manager catalog 只包含:
    • manager-skills/ 內全部 skills
    • regular-skills/ 中帶 shared tag 的 skills
  • 不要把其他 regular skills 混進 manager install flow
  • 只同步已經 finalize 完成的 canonical skills
  • 不要在這個 skill 內修改 canonical source
  • 若需要新增或修改 source,應改用 create-skill 或 update-skill

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.