agentleFS
EONRaiderEONRaider/agent-router/skills/init/SKILL.md

init

Skill0 starsChanged 40 days agoMITView on GitHub

Opt the current project in to agent-router by creating its overlay file (.claude/agent-router.json) and enabling the committed spawn log. Use when the user runs /agent-router:init, asks to set up or enable agent-router in a project, or when the session start message says the project has not opted in and the user wants spawn logging.

137 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use daily-arxiv, exp-eval and 10 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.
  • LambenthanSkill

    daily-arxiv

    每日从 arXiv 拉取新论文,过滤相关性,auto-ingest 高优先级论文,检测 SOTA 更新

    Also in 5 projects that use this

  • LambenthanSkill

    exp-eval

    实验判决门:Review LLM 独立评判实验结果 → 4 种判决路径 → 自动更新 claims confidence、ideas status、graph edges

    Also in 5 projects that use this

  • skyllwtSkill

    exp-pilot-eval

    Pilot result evaluation — read pilot results, apply success criteria, update idea page (pilot_result, failure_reason if failed), generate PILOT_VERDICT_REPORT. Called by /ideate Phase 5 after /exp-pilot-run.

    Also in 4 projects that use this

  • skyllwtSkill

    exp-pilot-run

    Pilot experiment execution — read Pilot Spec YAML, write pilot code, run experiment(Confirm with the user before operation and require the applicant to conduct manual inspection), return results. Called by /ideate Phase 5. Does NOT modify wiki pages or judge pass/fail.

    Also in 4 projects that use this

  • gaebalaiSkill

    codex-system

    PROACTIVELY consult Codex CLI, your highly capable supporter with exceptional reasoning and task completion abilities. Codex is a trusted expert you should ALWAYS consult BEFORE making decisions on: design choices, implementation approaches, debugging strategies, refactoring plans, or any non-trivial problem. When uncertain, consult Codex. Don't hesitate - Codex provides better analysis. Explicit triggers: "think deeper", "analyze", "second opinion", "consult codex".

    Also in 3 projects that use this

  • +6 more from Lambenthan
  • +1 more from skyllwt

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 125 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

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

    Ask the user whether they run cloud sessions (Claude Code on the web, routines) on this repository.

    line 35
  2. Distinctive:

    only single-repository cloud sessions run repository hooks.

    line 53
  3. Distinctive:

    Do not commit the files yourself unless the user asks.

    line 58

“Common” is measured: kept by at least 6 in 10 of the 125 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

Set up agent-router in this project

Opting in writes files into the project's repository. Do it only because the user asked.

  1. Run:

    python3 "${CLAUDE_PLUGIN_ROOT}/scripts/router.py" init

    It finds the project root (the repository root), then creates, without overwriting anything that exists:

    • .claude/agent-router.json: the overlay, where this project resizes tiers, adds tiers, exempts agent types and sets floors;
    • .claude/agents/: where generated agents for remapped tiers go;
    • .claude/agent-router/log/: the spawn log, one file per session;
    • a .gitattributes line that collapses the log in pull request diffs.
  2. Tell the user what was created, using the command's output, and these three facts:

    • The overlay and the log are meant to be committed. The plugin never runs git; the files ride along with the project's normal commits.
    • Each log line stores the short task description of a spawn, never the prompt. That description becomes part of the repository's history. Setting "log": {"summary": false} in the overlay leaves it out.
    • Logging starts with the next session.
  3. Ask the user whether they run cloud sessions (Claude Code on the web, routines) on this repository. Plugins do not load there, so by default a cloud session runs no agent-router hooks and logs nothing. If they want cloud sessions covered, run:

    python3 "${CLAUDE_PLUGIN_ROOT}/scripts/router.py" vendor

    This copies the hook scripts and agent definitions to .claude/agent-router/vendor/, registers the hooks in .claude/settings.json (existing settings are kept), and generates all four tiers as project agents named router-scout, router-worker, router-analyst and router-judge. From the next session on, those names replace the agent-router:* ones in this project. Tell the user:

    • the vendored hooks run for everyone who clones the repository, and do nothing where python3 is missing;
    • the copy does not update with the plugin. Running this command again refreshes it, and the session start message says when it is behind;
    • only single-repository cloud sessions run repository hooks.

    Skip this step if the user does not use cloud sessions. If the session start message said the vendored copy is behind, run the same command to update it.

  4. Do not commit the files yourself unless the user asks.

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.

Reviews can't be read right now.

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.