agentleFS
Sign inStart free
IamLumiRayIamLumiRay/Agent-delegate/SKILL.md

agent-delegate

Skill4 starsChanged 15 days agoMITView on GitHub

Delegate bounded tasks to Antigravity, a separate Codex profile, DeepSeek Harness, or Claude Code when requested, or when independent review, parallel work, or target capabilities justify delegation. Resume conversations and verify results.

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. anyoneanderson/agent-skills★ 3same name
    2. dylanjbarth/agent-skills★ 1same 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 repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    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:

    Keep coordination with the caller.

    line 16
  2. Distinctive:

    MCP available: use its five task tools with mcp.md.

    line 25
  3. Distinctive:

    Verify the target's identity and workspace using the provider's checks.

    line 30
  4. Distinctive:

    Default to read-only work.

    line 31
  5. Distinctive:

    Keep mappings and receipts local.

    line 37

≈ “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

Agent Delegate

Decide whether delegation helps, select an appropriate target, and verify its result.

Delegation decision

  • Handle simple or tightly coupled work locally. Delegate self-contained work when independent verification, parallel progress, or a known target capability outweighs startup and coordination cost.
  • Honor explicit provider, account, model, and delegation preferences. Otherwise, choose among registered targets using known capabilities, workspace access, and current availability; use provider checks instead of inferring capability from a model name.
  • Stay within the user's authorized scope and budget. If no suitable target is available, continue locally when feasible; report the limitation when delegation is required.
  • Keep coordination with the caller. Include caller/target context in the task brief; do not route work back along its delegation chain. Duplicate work only for deliberate independent checks, and retry failed delegation only when new evidence warrants it.

Select the provider

  • Antigravity / Gemini: antigravity.md.
  • Codex profile: codex.md. Runs a separate app-server under the selected profile, without attaching to its desktop window.
  • DeepSeek Harness: deepseek.md. Uses the installed SDK CLI and native login, with an owned runtime for each task.
  • Claude Code: claude.md. Uses the signed-in CLI in restricted print mode, with exact-session continuation.
  • First use: register the target with setup.md, then reuse its mapping.
  • MCP available: use its five task tools with mcp.md; otherwise use the helper scripts below.

Workflow

  1. Define the authorized workspace, deliverable, permitted actions, and stopping condition.
  2. Verify the target's identity and workspace using the provider's checks. Preserve the selected provider, account, and model.
  3. Default to read-only work. For authorized edits, use disjoint files or an isolated worktree.
  4. Save the conversation ID and receipt location. Continue with send; wait within the requested limit and require confirmed completion.
  5. Retrieve the visible final answer, verify relevant calculations or changes, and report the result and material limitations.

Entry point: scripts/invoke_agent.ps1, or the provider helper directly.

Keep mappings and receipts local; leave authentication to the provider application.

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