handoff
Skill82 starsChanged 42 days agoMITView on GitHub
Create or update a durable project handoff for the next AI agent/session. Use when ending a coding/research session, switching agents, compressing context, or preparing another Claude/ChatGPT/Codex session to continue the work.
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 repository27 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- ai-reviewSkill
- brieflySkill
- commit-batchingSkill
- commit-batchSkill
- create-alertSkill
- disclose-projectSkill
- discloseSkill
- evolve-claude-mdSkill
- figure-reviewSkill
- grillingSkill
- grill-meSkill
- lit-reviewSkill
- overbakedSkill
- pathfinderSkill
- pickupSkill
- recover-sessionsSkill
- repo-initSkill
- retroSkill
- reviewer-2Skill
- slack-messageSkill
- spot-aiSkill
- tab-setupSkill
- unstaleSkill
- worklogSkill
- write-new-skillSkill
The file
We split this skill into the rules it gives an agent
8 rules. 0 are common, 2 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 172 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 11
Create or update .ai/HANDOFF.md so the next agent can resume without rereading this conversation.
- Very specific: line 33
Read these if they exist: .ai/HANDOFF.md, CLAUDE.md, AGENTS.md.
- Very specific: line 39
Create .ai/ if missing. Append .ai/ to .gitignore if not already present.
- Distinctive: line 41
Use only the sections that have real content.
- Distinctive: line 76
Ask me at most one question, only if it would materially improve the handoff.
- Very specific: line 80
If this session produced something meant to be shared (a doc, Slack message, post, figure, published artifact) with no matching entry in …
- Very specific: line 87
After .ai/HANDOFF.md is written, run the worklog skill to capture this session to its three targets (local .ai/ mirror, remote server …
- Very specific: line 95
Run the evolve-claude-md skill to promote any durable session knowledge into the project's canonical agent file (CLAUDE.md, or the …
“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
Create or update .ai/HANDOFF.md so the next agent can resume without rereading this conversation.
Current repo state
git status --short 2>/dev/null || echo "(not a git repo)"git log --oneline -12 2>/dev/null || echo "(no git log)"git diff --stat HEAD 2>/dev/null | head -30 || echo "(no diff)"claude --version 2>/dev/null || echo "(claude --version unavailable)"Before writing
Read these if they exist: .ai/HANDOFF.md, CLAUDE.md, AGENTS.md. Scan recently modified files for TODO/FIXME markers. Do not invent results for commands that were not run.
Do not duplicate content already captured in durable artifacts (specs, plans, ADRs, issues, commits, diffs, CLAUDE.md). Reference them by path or URL and summarize only what the next agent needs to act. This includes review reports archived under .ai/reviews/ — point at them by path (e.g. in Next actions or Pitfalls) rather than restating findings. Redact anything sensitive — API keys, tokens, passwords, PII — even though .ai/ is gitignored.
Write .ai/HANDOFF.md
Create .ai/ if missing. Append .ai/ to .gitignore if not already present.
Use only the sections that have real content. Skip empty ones.
Objective
<!-- one sentence -->Next actions
<!-- numbered, concrete, specific — most important section; each action must be immediately executable without a follow-up question. Where a skill or subagent is the right tool, name it in the action itself (e.g. `/grilling` before a state-changing step; the stats-reviewer subagent once the estimator is drafted); point at `/pathfinder` if the next agent won't know which to reach for. --><!-- **Good**: `Run scripts/validate.py --baseline 1991-2020 and check the NaN count in coastal cells` --><!-- **Good**: `Run /grilling on the resampling design before editing pipeline.py` --><!-- **Bad**: `Continue working on the validation` -->State
<!-- branch, environment, anything non-obvious; files changed and why --><!-- If the session established harness-behavior facts (parser, hooks, skill
loading), record the `claude --version` from the preamble — those facts are
version-scoped. Register: docs/harness-behavior.md in ai-science-toolkit. -->Decisions
<!-- non-obvious choices made this session: what, why, what was rejected -->Scientific context
<!-- datasets, versions, paths, provenance; baselines and reference periods;
counterfactual definitions; methodological assumptions;
known data issues: NaNs, masking, coordinate alignment, scaling -->Pitfalls
<!-- what already went wrong, or is likely to -->Open questions
<!-- only if genuinely unresolved and consequential: the question,
what breaks if answered differently, best current recommendation -->Preserve useful prior content. Remove stale or superseded details. Distinguish facts, assumptions, and recommendations explicitly.
Ask me at most one question, only if it would materially improve the handoff.
Unrated shareables
If this session produced something meant to be shared (a doc, Slack message, post,
figure, published artifact) with no matching entry in .ai/disclosures.jsonl, list
it in one line in your reply and suggest running the disclose skill (/disclose). Don't run it
from here — the rating is interactive and belongs to the artifact, not the session.
Ship a worklog entry
After .ai/HANDOFF.md is written, run the worklog skill to capture this session
to its three targets (local .ai/ mirror, remote server cache, Notion weekly
page). Pass it the session summary and next actions you wrote above so it does not
re-derive them. worklog is best-effort by design — it must never block or fail
the handoff.
Finally
Run the evolve-claude-md skill to promote any durable session knowledge into the project's canonical agent file (CLAUDE.md, or the AGENTS.md it redirects to — the skill resolves which).
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.