handoff-contract
Skill1 starsChanged 4 months agoMITView on GitHub
Write a self-contained handoff when work moves to a new session, a different AI tool, or another person. Use when context is about to be lost — quota exhausted, switching tools, ending a session mid-project.
In this repository14 other files it gives its agents.
- bad-news-firstSkill
- bug-autopsySkill
- context-hygieneSkill
- decision-wiringSkill
- dispatch-economicsSkill
- grill-founderSkill
- loop-contractSkill
- mainlineSkill
- model-tieringSkill
- red-light-firstSkill
- rules-as-mechanismSkill
- surgical-changesSkill
- two-axis-reviewSkill
- vertical-ticketsSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
Write facts, not vibes. Every claim verifiable, every exit observable.
≈ “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.
Files
The receiver knows nothing. Not your conversation, not your decisions, not your file layout. A handoff that assumes shared memory silently loses work — we watched a receiving tool merge a pull request into a dead branch because the handoff skipped one re-review rule, and the work vanished for days before anyone noticed.
A handoff must contain
- Status snapshot — what is done, and how it was verified (test counts, CI links). Not "mostly finished".
- Rules of the road — the 5–8 constraints that actually bind (what never to touch, what needs approval, what "tests may not be weakened" means here). Not your whole culture doc.
- Environment recipe — exact commands to reach a working state, copy-pasteable.
- Task list with per-task exit conditions — each task states its own observable "done".
- Overall exit — when the whole handoff is complete, in checkable terms.
- Not-yours list — parked items and escalation paths, so the receiver doesn't "helpfully" wander into them.
- File map — where the five things they'll look for actually live.
Write facts, not vibes. Every claim verifiable, every exit observable.
Success criteria
The stranger test: someone with zero context could execute the handoff end to end. If any step requires "ask the previous session", the handoff failed.
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
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.

