handoff
Skill6 starsChanged 8 months agoMITView on GitHub
Save a structured handoff document so another Claude instance (or this one after /clear) can resume the work. Use when context is getting long, switching tasks, or before /clear.
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 repository5 other files it gives its agents.
- avadakedavraSkill
- obliviateSkill
- pensieveSkill
- replaySkill
- timeturnerSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 3 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 172 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Create a structured handoff document that any Claude instance can read to resume this work.
- Distinctive: line 19
Test status (passing/failing)
- Very specific: line 78
If there's an active task list, pass all pending tasks via --tasks
- Distinctive: line 79
Do NOT include raw code dumps — reference files and line numbers instead
- Very specific: line 81
Only include files YOU modified this session in --files — not the user's other work
“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
Session Handoff
Create a structured handoff document that any Claude instance can read to resume this work.
Steps
- Gather state — Review the current conversation to understand:
- What task was requested
- What has been completed so far
- What remains to be done
- Key decisions made and why
- Files created or modified (collect the full list)
- Current blockers or open questions
- Test status (passing/failing)
- Approaches tried that failed (with error messages)
- Any task list items (check /tasks)
- Active skill/workflow — If a skill was active (e.g.,
/plan-guarded,/bugfix,/refactor), note which skill and which phase/step the session was in
Write the handoff — Pipe a markdown document to
handoff-create.sh. The script handles all path computation, git patches, YAML frontmatter, and team copies automatically. Always pass--session-idwith$CLAUDE_SESSION_IDso the handoff can be found by auto-resume after/clear.REMEMBRALL_ROOT="${CLAUDE_PLUGIN_ROOT:-$(cat /tmp/remembrall-meta/plugin-root 2>/dev/null)}" cat << 'HANDOFF_CONTENT' | bash "${REMEMBRALL_ROOT}/scripts/handoff-create.sh" \ --cwd "$(pwd)" \ --session-id "$CLAUDE_SESSION_ID" \ --status "in_progress" \ --files "file1.ts,file2.ts,file3.ts" \ --tasks "Remaining task 1" "Remaining task 2" "Remaining task 3" # Session Handoff **Task:** [One-line description of what was requested] **Active Skill:** [Skill name if one was active, e.g., `/plan-guarded` Phase 2 step 3, `/bugfix` step 4 — or "none"] ## Completed - [Bullet list of what's done, with file paths] ## Next Step — Do This First [The EXACT next action. Not a list of everything remaining — just the single next thing to do. Be specific: "Run the test suite for auth module" not "continue testing". Include the file path and line number if applicable.] ## Remaining (after next step) - [Numbered list of what still needs to happen, in priority order] ## Failed Approaches - [Approaches tried and ruled out. For each: what was attempted, the exact error message or why it failed, and why it won't work. This prevents the next session from repeating the same dead ends.] ## Do NOT Do - [Things the next session must NOT do — files to leave alone, features that are done and should not be re-analyzed, rabbit holes to avoid] ## Key Decisions - [Important choices made and why — so the next instance doesn't re-debate them] ## Files Modified - [List of files changed in this session, with brief note of what changed] ## Context [Any important context the next instance needs — error messages, gotchas discovered, user preferences expressed during the session] ## Open Questions - [Anything unresolved that needs user input] HANDOFF_CONTENTReplace the placeholder content with actual information from this session. Replace the
--filesand--tasksarguments with real values.- Confirm and continue — Then:
- Autopilot mode (REMEMBRALL autonomous flag is active, or AUTONOMOUS MODE was mentioned in a recent nudge): Do NOT enter plan mode. Do NOT ask the user anything. Just say "Handoff saved" in one line and keep working on the next task. The handoff is a safety net — compaction will happen naturally and the session-resume hook will auto-inject it. This is the autopilot loop: save → keep working → compaction → auto-resume → keep working.
- Attended mode, context is low (triggered by REMEMBRALL_WARN or REMEMBRALL_URGENT): call
EnterPlanModeimmediately. The session-resume hook will auto-inject the handoff after context clears. Do NOT tell the user to type/clearor/replay. - Manual invocation (context is fine): just confirm with a brief summary and continue working.
Rules
- Be concise but complete — the next instance has zero context
- Include file paths, not vague descriptions
- If there's an active task list, pass all pending tasks via
--tasks - Do NOT include raw code dumps — reference files and line numbers instead
- The handoff must be self-contained: another Claude reading only this file should be able to continue
- Only include files YOU modified this session in
--files— not the user's other work - Set
--statustoblockedif work is stuck,pausedif intentionally stopping
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.

