agentleFS
Sign inStart free
hcsolakogluhcsolakoglu/sidekick/skills/claude-code/SKILL.md

sidekick

Skill1 starsChanged 4 months agoMITView on GitHub

Orchestrate persistent Codex, Devin, Claude Code, Hermes, or mock workers with the sidekick CLI. Use for delegated work that should run concurrently, survive the parent turn, resume by session, or be cancelled and harvested reliably.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. sksareen/sidekick★ 1same name
    2. specscore/specstudio-skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository4 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    Run sidekick doctor ENGINE before first use on a machine.

    line 8
  2. Very specific:

    Start one background sidekick wait NAME...

    line 11
  3. Very specific:

    Use sidekick adopt ENGINE NAME --session ID --dir PATH for an existing session.

    line 15

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Sidekick orchestration

  1. Run sidekick doctor ENGINE before first use on a machine.
  2. Inspect sidekick capabilities ENGINE --model MODEL --json before choosing model, provider, transport, effort, permission, sandbox, or workspace trust.
  3. Delegate a bounded task with sidekick spawn ENGINE NAME --dir PATH -- "PROMPT". Use unique, stable names and separate control flags only when the evidence matrix marks them native.
  4. Start one background sidekick wait NAME... --all --json process for the batch. Continue useful parent work while it waits; do not start polling loops.
  5. Harvest with sidekick result NAME --json. Check both status and exitCode before trusting output.
  6. Continue the same harness session with sidekick send NAME -- "FOLLOW-UP".

Use sidekick adopt ENGINE NAME --session ID --dir PATH for an existing session. Use sidekick tail NAME for live output and sidekick cancel NAME to stop the full process tree while keeping the session resumable. Use send --force only when replacing an active turn.

Sidekick persists requested, applied, and effective control observations. An argv flag does not prove provider-reported effective state. Unsupported, malformed, model-dependent-unverified, or runtime-version-mismatched explicit values fail closed with exit 2. Codex effort is value-gated (minimal|low|medium|high|xhigh|max) for any model. Claude effort is exact-model; inspect capabilities before requesting it. Hermes has no public per-run effort flag. Devin family slug glm-5.2 and variant UID glm-5-2 are distinct and are never silently aliased; do not use --mode high as effort.

Choose an engine by task and local availability: use Codex or Claude for focused repository work, Devin for broad autonomous implementation, Hermes for general tool-rich work, and mock only for deterministic workflow tests. Split independent tasks; keep tightly coupled edits with one worker. Give every delegate an exact scope, constraints, verification command, and expected response.

Do not concurrently spawn workers with the same --dir: harness session discovery and workspace state can be directory-scoped, causing contention, accidental serialization, or overlapping edits. For real parallelism, create a separate git worktree for each worker and pass its path through --dir. If the tasks must share one working tree, serialize them and reuse the existing run with sidekick send.

Prompt transport is handled by Sidekick: Codex and Claude use stdin, Devin uses --prompt-file, and Hermes currently uses argv. Hermes prompts near 32 KB are unsafe on native Windows; prefer WSL or another engine there. See sidekick doctor hermes.

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.

Check your setup, free