Claude handoff
Skill dotknewt/awesome-agency/skills/in-progress/claude-handoff
Use when the user asks to hand off, delegate, or continue the current work in a background or new session — e.g. "can you keep working on this while I step away", "spin up a background agent for this", "hand this off", "continue this without me", "keep going on this while I'm in a meeting" — even if they don't say "handoff" explicitly. Writes a redacted, artifact-referencing handoff summary and launches a fresh background agent seeded with it via `claude --bg`.From its SKILL.md
npx -y skills add dotknewt/awesome-agency --skill claude-handoffAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
3.7 KB, as published. Nobody here has run it
Write a handoff summary of the current conversation so a fresh agent can continue the work. Instead of saving it, launch a background agent seeded with the summary as its prompt: claude --bg --name "<descriptive name>" "<handoff summary>". It starts in the current working directory and returns immediately; the user manages it with claude agents.
Always pass -n/--name with a descriptive name (e.g. --name "Fix login bug") — it sets the display name shown in the job list, session picker, and terminal title. If the user gave a name (e.g. "spin up a background agent named migrate-db"), use it verbatim.
If the user passed arguments (e.g. /claude-handoff focus on the auth refactor), treat them as a description of what the next session will focus on and tailor the summary accordingly.
Handoff Summary
Structure the summary with this template so output is consistent across handoffs — adapt the depth of each section to the task, but keep all five headings:
## Handoff Summary
### Context
[Why this work is happening — the original request or goal, one to three sentences.]
### Current State
[What's done, what's in progress, and where things stand right now. Point at files and paths rather than pasting their contents.]
### Next Steps
[Ordered list of what the new agent should do next.]
### Suggested Skills
[Skills the new agent should invoke and why, e.g. "- `verify` — confirm the migration script runs end to end".]
### References
[Paths or URLs to PRDs, plans, ADRs, issues, commits, or diffs relevant to the work. Do not paste their contents — see Gotchas.]
Pass this whole markdown block as the <handoff summary> argument to claude --bg.
Gotchas
- Never inline a full file diff, PRD, or command output dump in the summary — reference the path (and line range, if relevant) instead. The summary becomes the new agent's prompt; a wall of pasted content burns its context before it's read anything.
- Before including any command output in the summary, scan it for API keys, tokens, passwords, and other secrets (common patterns:
sk-,ghp_,AKIA,Bearer,-----BEGIN,.envcontents, connection strings with embedded credentials). Redact or omit rather than guess — the summary becomes the agent's prompt and may be logged or displayed. - Don't duplicate content already captured in other artifacts (PRDs, plans, ADRs, issues, commits, diffs). Reference them by path or URL in the References section instead of restating them.
- The
--namevalue is a display label only — it doesn't change the working directory or the session ID. Don't rely on it to disambiguate jobs programmatically.