handoff
Skill4 starsChanged 11 days agoMITView on GitHub
Write a curated, size-capped handoff file that carries task state from this session to a fresh one. Run only when the user invokes /handoff.
191 other projects use a version of this.
Adopted by
Most recently taken up by
- xiaoxiaosrm/dsh-mattpocock-skillstheir own version, first seen Nov 8, 2026
- ethosagent/ethostheir own version, first seen Nov 8, 2026
- dsifry/metaswarmtheir own version, first seen Nov 8, 2026
- glebis/claude-skillstheir own version, first seen Nov 8, 2026
- domengabrovsek/agent-configtheir own version, first seen Nov 8, 2026
- domengabrovsek/claudetheir own version, first seen Nov 8, 2026
Used with
Projects with it also use decision-mapping, setup-ts-deep-modules and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
FabienDehopreSkilldecision-mapping
Turn a loose idea into a sequenced map of investigation tickets, then drive them to resolution one at a time.
Also in 8 projects that use this
liufeng1976Skillsetup-ts-deep-modules
Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.
Also in 42 projects that use this
AWSSkillto-questionnaire
Turn a decision you can't fully answer into a questionnaire for someone else to fill in.
Also in 93 projects that use this
GoogleSkillask-matt
Ask which skill or flow fits your situation. A router over the skills in this repo.
Also in 87 projects that use this
RoentekSkillloop-me
Grill me about specs for the workflows I want to build, within this workspace.
Also in 51 projects that use this
LesaniSkillwriting-beats
Shape an article as a journey of beats, choose-your-own-adventure style. The user picks a starting beat from the raw material, you write only that beat, then offer options for where to pivot next, beat by beat, until the article reaches a natural end. Use when the user has raw material and wants to assemble it as a narrative rather than an argument.
Also in 54 projects that use this
Other versions191 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 191
- mattpocock/skills★ 309ksame name
- QuivrHQ/quivr★ 40ksame name
- The-Vibe-Company/quivr★ 40ksame name
- GreyDGL/PentestGPT★ 16ksame name
- ykdojo/claude-code-tips★ 10ksame name
- ArcReel/ArcReel★ 6.3ksame name
- vinvcn/mattpocock-skills-zh-CN★ 5.5ksame name
- withcoral/coral★ 4.9ksame name
- epicenter-md/epicenter★ 4.8ksame name
- EpicenterHQ/epicenter★ 4.8ksame name
181 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository24 other files it gives its agents.
- AGENTS.mdAGENTS.md
- skills/client-config-opencode/references/agents.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- repository-reviewSkill
- weekly-skill-refreshSkill
- skill-maintenanceSkill
- about-me-skill-creatorSkill
- agent-creator-claudecodeSkill
- agent-creator-codexSkill
- agent-creator-copilotSkill
- agent-creator-opencodeSkill
- client-config-claudecodeSkill
- client-config-codexSkill
- client-config-copilotcliSkill
- client-config-opencodeSkill
- copilot-instruction-creatorSkill
- copilot-workerSkill
- docker-image-managementSkill
- git-commit-messagesSkill
- google-style-docsSkill
- markdownlint-validatorSkill
- mkdocs-site-managementSkill
- skill-routerSkill
The file
We split this skill into the rules it gives an agent
9 rules. 0 are common, 6 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 172 edited versions we read keep these.
6
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 10
Write one live file, .agent-output/handoff.md, so a fresh session can pick up this task where it stopped.
- Distinctive: line 17
Run this before the first write, from inside the repository:
- Distinctive: line 24
Exit 1 (not ignored): stop and ask which the user prefers.
- Very specific: line 26
A. Create .agent-output/ and add .agent-output/ to the repository's .gitignore.
- Very specific: line 42
Create .agent-output/ if it does not exist, then write the file in this shape.
- Distinctive: line 69
Keep the whole file to about 100 lines.
- Distinctive: line 70
Leave out anything a fresh session can derive from git log, git diff, or the code.
- Distinctive: line 71
Leave out durable facts and preferences.
- Distinctive: line 72
Never include credentials, tokens, or secrets.
“Common” is measured: kept by at least 6 in 10 of the 172 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Handoff
Write one live file, .agent-output/handoff.md, so a fresh session can pick up this
task where it stopped. Overwrite any previous handoff. The file holds task state only,
not durable facts: those belong in the harness's memory feature.
1. Check that the path is ignored
A handoff can hold proprietary context, so never write it to a path git would stage. Run this before the first write, from inside the repository:
git check-ignore -q .agent-output/handoff.md- Exit 0 (ignored): continue to step 2.
- Exit 1 (not ignored): stop and ask which the user prefers. Write nothing until they
answer.
- A. Create
.agent-output/and add.agent-output/to the repository's.gitignore. This is a tracked change that every contributor receives. - B. Append
.agent-output/to the file printed bygit rev-parse --git-path info/exclude. This is untracked and applies to this clone only.
- A. Create
- Not a git repository (the command reports that): continue to step 2. There is nothing to ignore.
2. Gather the header facts
- Timestamp:
date -u +%Y-%m-%dT%H:%M:%SZ - Branch:
git symbolic-ref --short -q HEAD. If that prints nothing, use(detached). Outside a git repository, use(none).
3. Write the file
Create .agent-output/ if it does not exist, then write the file in this shape. The
header lines must keep this exact form, because the session-start hook reads the
Branch: line.
# Handoff
Written: <timestamp>
Branch: <branch>
## Goal and current state
## Decisions
## Gotchas and dead ends
## Next step
## In-flight work
## Open questions- Goal and current state: what the task is and how far it got.
- Decisions: each decision with the reason for it.
- Gotchas and dead ends: approaches already tried and why they failed.
- Next step: the exact next action, specific enough to start without re-deriving it.
- In-flight work: uncommitted or half-done changes, by file.
- Open questions: what still needs an answer, and from whom.
Rules:
- Keep the whole file to about 100 lines. Cut to fit.
- Leave out anything a fresh session can derive from
git log,git diff, or the code. - Leave out durable facts and preferences. They belong in memory, not here.
- Never include credentials, tokens, or secrets.
- Omit a section only if it is truly empty. Write
Nonerather than inventing content.
4. Report back
Tell the user the path, the line count, and that the file is meant to be reviewed before
use. Suggest /clear when they are ready to continue in a fresh session. Do not run it
for them.
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.

