Claude Code Handoff and Resume Skills

# Claude Code Handoff & Resume Skills (setup prompt)

Paste everything in the block below into Claude Code. It builds two skills:
`handoff` (save session state) and `resume` (pick it back up). They work in any
project, with no external tool, vault, or database required.

---

```
Create two Claude Code skills for me: "handoff" and "resume". They work as a pair.
handoff saves the state of a working session to a file, resume reads that file back
and gets us going again. Build them to be portable and self-contained, with no
dependency on any external tool, vault, database, or service.

Put each skill in its own folder under .claude/skills/ in the current project
(handoff/SKILL.md and resume/SKILL.md). Use this design:

=== SKILL 1: handoff ===

Description / triggers: invoke when the user says "handoff", "/handoff", "wrap",
"/wrap", "save the session", "end session", "hand this off", or "let's stop here".
Also suggest it proactively when a session has run long or just finished a big task,
but only suggest, don't auto-run it.

What it does when invoked:
1. Write a single handoff file to .handoff/latest.md in the project root (create the
   .handoff/ folder if it doesn't exist). Also write a timestamped copy to
   .handoff/archive/<YYYY-MM-DD-HHMM>.md so history is preserved.
2. The file must capture everything a fresh session needs to continue with zero
   re-explaining. Use this structure:

   # Handoff <project name> <date/time>
   ## What we were doing
   One short paragraph: the goal of this session.
   ## State right now
   What's done, what's in progress, what's verified vs unverified.
   ## Files touched
   Bullet list of files changed/created this session, each with a one-line note.
   ## Next steps
   Ordered list. The first item is the single most important next action.
   ## Open questions / blockers
   Anything waiting on a decision or an external thing.
   ## How to verify / run
   Exact commands to build, test, or run the project, if relevant.
   ## Notes
   Anything else that mattered but doesn't fit above.

3. Keep it tight and factual, with no filler. Only write what's true and verified. If
   something wasn't tested, say so explicitly.
4. After writing, tell the user the file path and give a 3-line summary of what was
   saved.

=== SKILL 2: resume ===

Description / triggers: invoke when the user says "resume", "/resume", "pick up where
we left off", "continue", "where were we", or "/resume <topic>".

What it does when invoked:
1. Read .handoff/latest.md. If it doesn't exist, say so plainly and offer to start
   fresh. Don't guess.
2. If the user passed a topic and there are multiple archived handoffs, scan
   .handoff/archive/ for the most relevant one and use that instead.
3. Synthesize the handoff into a short orientation brief (max ~10 lines): what we
   were doing, the current state, and the single next action, ending with
   "Ready when you are."
4. Do NOT start doing work automatically. Just orient and wait for the go-ahead,
   unless the user clearly told you to continue immediately.

=== Both skills ===

- Write the SKILL.md frontmatter (name + description) so the triggers above fire
  reliably.
- Keep the skills plain and universal so they work in any repo or project type.
- After creating both, show me the two file paths and a one-line description of each,
  then do a quick self-check that the trigger phrases don't collide.
```

---

## Notes

- Skills live in `.claude/skills/handoff/SKILL.md` and `.claude/skills/resume/SKILL.md`.
- The handoff file format is identical to the Codex version. A `.handoff/latest.md`
  written by Claude can be read by Codex and vice versa.
- For a global version (carry handoffs across every repo), tell Claude to write to
  `~/.claude/handoffs/` instead of the per-project `.handoff/` folder.