Install
$ agentstack add skill-toolmonsters-handoff-skill-handoff-skill ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo 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.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
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 →About
Handoff: continue this conversation in any LLM
You are about to be replaced mid-task. Your job: write a handoff document so complete that the next model (ChatGPT, a fresh Claude session, anything) continues the work seamlessly, without asking the user a single question about what happened here.
Output
Produce ONE markdown document, nothing else. If the platform lets you create a file or artifact, make it a file named handoff-[topic].md. Otherwise output it in a single markdown code block so the user can copy it in one click.
Structure (all sections required)
# Handoff: [one-line topic]
## Context & goal
Who the user is (only what this conversation revealed), what they are trying to
accomplish, and why. 3-5 lines.
## Key decisions made
Every choice that was settled, with the reasoning if it was stated.
Bullet list. These must NOT be re-litigated by the next model.
## Current state
Where the work stands right now. What is done, what is validated,
what is mid-flight.
## Work in progress (verbatim)
The exact current version of whatever is being worked on: the full draft,
the code, the list, the plan. Paste it COMPLETE, word for word.
This is the most important section. The next model must not reconstruct
or re-generate it from a description.
## Rejected paths
What was tried and abandoned, and why. Prevents the next model from
proposing them again.
## Tone & working preferences
How the user likes to work, phrased as instructions
(e.g. "short answers", "always show the full text", "no em-dashes",
"push back when data contradicts the idea").
## Immediate next step
The single thing the next model should do first, stated as an instruction.
Strict rules
- Zero invention. Only facts, decisions and text that literally appear in this conversation. If something is unclear, write "unclear" rather than guessing.
- Verbatim beats summary. For anything in progress (a draft, a prompt, code), include the full exact text, never a description of it.
- Keep every number, name, file path, URL and date exactly as they appeared.
- Write it for a stranger. The next model knows nothing: no internal shorthand, no "as discussed above".
- No preamble, no sign-off. The document starts at
# Handoff:and ends at the next step.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: ToolMonsters
- Source: ToolMonsters/handoff-skill
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.