AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Reboot

skill-ao92265-claude-code-playbook-reboot · by ao92265

Distill the current task and state into a clean, self-contained reprompt to paste into a fresh session, so a stale or bloated context can be cleared without losing the thread. Triggers "/reboot", "reboot this session", "this session is stale", "stale session", "clear and reprompt", "reprompt me", "fresh start", "this session's getting old". Do NOT use for a cross-terminal morning digest (that's m…

No reviews yet
0 installs
28 views
0.0% view→install

Install

$ agentstack add skill-ao92265-claude-code-playbook-reboot

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-ao92265-claude-code-playbook-reboot)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Reboot? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

reboot

The session is old or bloated and should be cleared. Your job: hand the user a clean, self-contained prompt they can paste into a fresh session and keep going — no thread lost.

You can't run /clear yourself. You produce the reprompt; the user clears and pastes.

Sources

  • Primary: the current conversation you already hold. That's the real source of the goal and state.
  • Cross-check / fill gaps: the freshest handoff in ~/.claude/handoffs/ — pick the one matching the current working directory, else the most recently modified — and its .compact.md sibling if present. These are the same sources sessionstart-handoff.sh re-injects into a new session.

Distill into the reprompt

Capture only what a fresh Claude needs:

  • Goal — what "done" looks like, in one or two lines.
  • State — what's already done / decided.
  • Next step — the immediate thing to do.
  • Constraints — hard rules or decisions already locked in.
  • Key paths — files/dirs in play.

Drop dead-ends, resolved tangents, and tool noise.

Output

Two short lines first: why the session looks stale, and what to do — /clear (or open a new terminal), then paste the block below.

Then one fenced code block = the paste-ready reprompt. Write it as a direct instruction to a fresh Claude. It must be fully self-contained — assume the reader saw none of this session. No "as discussed", no dangling pronouns.

Honesty

Don't invent state. If branch, dirty files, or anything else is uncertain, write "verify before acting" instead of asserting it — same disclaimer the handoff itself uses. A wrong "current state" is worse than an admitted gap.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.