agentleFS
MathCazMathCaz/claude-skills/handoff-M/SKILL.md

handoff-M

Skill1 starsChanged 4 months agoMITView on GitHub

Hand off either the full conversation OR a single topic/sub-thread to another agent. Use `full` to compact the entire session for a fresh agent (current session typically ends). Use `topic` to extract one sub-thread (a bug spotted on the side, a tangential idea) for another agent while you keep working in the current one.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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:

    Write a handoff document so another agent can pick up specific work.

    line 9
  2. Distinctive:

    Topic mode only: after producing the doc, return cleanly so the user can resume the main thread — do not narrate next steps for the current …

    line 34

≈ “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

Handoff

Write a handoff document so another agent can pick up specific work. Save to the OS temp directory — never the current workspace.

Modes

The first arg word selects the mode:

  • full <focus> — compact the entire conversation. Use when you're wrapping up this session and want a fresh agent to continue the work. The current session is typically ended after.
  • topic <topic description> — extract only the sub-thread matching the topic. Use when a side-issue surfaced (a bug, a tangential idea) that deserves its own agent while you stay on the main thread here. The current session continues — do not summarize unrelated content, do not "wrap up" the current work.

If no mode is given, ask the user which one. For topic without a description, ask which topic to extract — the skill cannot proceed without knowing. For full without a description, proceed with a comprehensive summary; a description is optional and only biases the doc's framing.

What goes in the document (both modes)

  • Goal — what the receiving agent should accomplish (verbatim from user args when provided)
  • Context — only what the receiving agent needs to be effective
    • Full mode: comprehensive — decisions made, current state, open items, blockers
    • Topic mode: focused — the topic's history within this conversation, what's known/unknown, current state of that thread, why it was set aside
  • Suggested skills — skills the receiving agent should invoke for this work
  • Next steps — concrete first moves the receiving agent can take

Rules (both modes)

  • Do not duplicate content already in PRDs, plans, ADRs, issues, commits, diffs — reference them by path or URL
  • Redact sensitive info — API keys, passwords, PII, internal secrets
  • Save to OS temp dir, print the absolute path back to the user
  • Topic mode only: after producing the doc, return cleanly so the user can resume the main thread — do not narrate next steps for the current session

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.