agentleFS
Sign inStart free
p10ns11yp10ns11y/skills/subagent-delegation/SKILL.md

subagent-delegation

Skill2 starsChanged 40 days agoMITView on GitHub

Delegate broad exploration to a subagent with a fixed Return format. Use when the task needs readonly repo mapping, token/CSS/prosody surveys, QA module exploration, or "explore thoroughly" with numbered deliverables.

12 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 12 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do not duplicate full exploration in the parent — synthesize and propose next steps

    line 31
  2. Distinctive:

    If subagent finds a small fix, parent implements after exploration returns

    line 32

“Common” is measured: kept by at least 6 in 10 of the 12 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

Subagent delegation

CLT: ../rules/clt-dual-load.mdc — parent stays light (extraneous ↓); subagent returns the fixed format only (no transcript dump); fan-out readonly explores when independent.

When to use

  • User lists numbered exploration goals (files, entry points, data flow, pain points)
  • Task is breadth-first across many files; parent session should stay lightweight
  • User says "explore in readonly mode" or references agent/subagent branches

Spawn contract

Give the subagent:

  1. Scope — directory or repo root (use {REPO_ROOT} in notes)
  2. Readonly — explore agents should not implement unless asked
  3. Return format (required):
    • File tree with one-line purpose per file
    • Mermaid or bullet flow (entry → core → output)
    • Pain points / readability notes
    • Direct answers to numbered questions

Parent session

  • Link subagent output back to the user’s original numbered list
  • Do not duplicate full exploration in the parent — synthesize and propose next steps
  • If subagent finds a small fix, parent implements after exploration returns

Tool pattern (from gold subagents)

Typical chain: UpdateCurrentStep → Glob → Grep → Read — no Write until parent approves implementation.

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