agentleFS
fagemxfagemx/edda/.claude/skills/coord-handoff/SKILL.md

coord-handoff

Skill36 starsChanged 2 months agoApache-2.0View on GitHub

Clean handoff when finishing multi-agent work — summarize, decide, unclaim

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    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. Distinctive:

    Before session end when peers will continue working

    line 13
  2. Distinctive:

    When switching to a different task and releasing your current scope

    line 14
  3. Distinctive:

    Run these commands to understand what changed:

    line 20
  4. Distinctive:

    Check for incomplete work:

    line 31
  5. Very specific:

    Search modified files for TODO and FIXME comments

    line 32

≈ “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

Coordination Handoff

You are a coordination specialist. Your role is to help the agent cleanly exit a multi-agent session by summarizing work, recording decisions, and preparing peers for continuation.

When to Use

  • Finishing a task in a multi-agent session
  • Before session end when peers will continue working
  • When switching to a different task and releasing your current scope

Workflow

Step 1: Summarize Changes

Run these commands to understand what changed:

git diff --stat
git log --oneline $(git merge-base HEAD main)..HEAD

Summarize: how many commits, which files, what was the goal.

Step 2: List Unfinished Items

Check for incomplete work:

  • Search modified files for TODO and FIXME comments
  • Review the task list for incomplete items
  • Note any partial implementations or known issues

Step 3: Record Decisions

For each architectural choice made this session that other agents need to know:

edda decide "<domain.key>=<value>" --reason "<why>"

Focus on:

  • Library/framework choices
  • API design decisions
  • Schema or data structure changes
  • Configuration choices

Skip: formatting changes, test additions, minor refactors.

Step 4: Post Session Note

Summarize the session for peers:

edda note "<summary>" --tag session

Include:

  • What was completed
  • What decisions were made
  • What remains for the next session or other agents

Step 5: Scope Release

Where the Claude Code hooks are wired, unclaim happens automatically on session end via the SessionEnd hook — nothing to do.

Where they are not (claims made from a bare CLI, CI, or a host without the bridge hooks), the claim outlives the session: there is currently no edda unclaim verb to release it by hand (GH-455). In that case, say so in your handoff summary — peers reading the board should know the claim is dead even though it still folds. Enforcement stays safe either way: every consumer joins claims against live heartbeats.

Output Format

Present the handoff summary:

## Coordination Handoff

### Changes Made
- N commits, M files modified
- <1-2 sentence summary of work done>

### Decisions Recorded
- <key> = <value> (reason)

### Unfinished Work
- <items remaining, with context for the next agent>

### Session Note
> <the note posted via edda note>

### Scope Released
- <label> will be unclaimed on session end (peers can then work in <paths>)

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.