agentleFS
sudeepgupta90sudeepgupta90/my-ai-marketplace/skills/handoff/SKILL.md

handoff

Skill0 starsChanged 9 days agoMITView on GitHub

Compress the current session into a handoff document so work can continue in a fresh session or move to a different agent. Use when the context window is filling up, when the user says to hand off, wrap up, continue later, or start fresh, or when switching to another tool or teammate mid-task.

191 other projects use a version of this.

Adopted by

Most recently taken up by

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.
  • FabienDehopreSkill

    decision-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

  • liufeng1976Skill

    setup-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

  • AWSSkill

    to-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

  • GoogleSkill

    ask-matt

    Ask which skill or flow fits your situation. A router over the skills in this repo.

    Also in 87 projects that use this

  • RoentekSkill

    loop-me

    Grill me about specs for the workflows I want to build, within this workspace.

    Also in 51 projects that use this

  • LesaniSkill

    writing-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

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is 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.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Write it to a file the next session can read — default docs/handoff-<topic>.md, or a path the user names.

    line 21
  2. Distinctive:

    Verify before asserting: Every file path and command in the handoff must be real.

    line 56
  3. Distinctive:

    State what is unverified: If tests were never run, say so.

    line 58
  4. Distinctive:

    Keep it short enough to be read: If it exceeds roughly two pages, the detail belongs in the codebase or a design doc that the handoff links …

    line 60

“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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Handoff

Produce a document that lets someone — a fresh session, a different agent, or a person — resume this work without re-deriving what was already established.

When to reach for this

  • Context is filling and the session will be summarized or restarted
  • The user is stopping for the day and wants to resume cleanly
  • Work is moving to another agent (Codex, Gemini) or another person
  • A long investigation reached a conclusion worth preserving beyond this session

What makes a handoff good

A handoff is not a transcript summary. It is the minimum a competent stranger needs to continue. Optimise for what would otherwise be re-derived at cost: decisions and their reasons, dead ends already ruled out, and the exact next action.

Write it to a file the next session can read — default docs/handoff-<topic>.md, or a path the user names. Do not print a long handoff into the conversation; the point is that it survives the conversation ending.

Structure

# Handoff: <topic>

**Status:** <one sentence: what is done, what is in flight>
**Updated:** <absolute datetime>

## Goal
What we are trying to achieve, and why. Absolute dates, no "yesterday".

## Current state
What exists now. Name real files and functions with paths. Distinguish
"built and verified" from "built but untested" from "planned".

## Decisions made
| Decision | Why | Alternatives rejected |
Enough that the next session does not reopen a settled question.

## Dead ends
What was tried and did not work, and the symptom. This is often the highest
value section: it is pure cost to rediscover.

## Next action
The single specific thing to do next, concrete enough to start immediately.
Then anything queued behind it.

## Open questions
Things genuinely undecided, and who or what can decide them.

Rules

  1. Verify before asserting. Every file path and command in the handoff must be real. A handoff that points at a function that does not exist is worse than none — the reader trusts it and loses time.
  2. Absolute dates. "Last Tuesday" is meaningless to a session that starts in three weeks.
  3. State what is unverified. If tests were never run, say so. Never imply completion that did not happen.
  4. Prefer pointers to prose. A path plus a line number beats a paragraph of description.
  5. Keep it short enough to be read. If it exceeds roughly two pages, the detail belongs in the codebase or a design doc that the handoff links to.

Cross-agent handoffs

When the next reader is a different tool, avoid Claude-specific vocabulary. Say "run the tests with npm test" rather than naming a slash command that only exists here. Skill and plugin names are Claude-specific; describe the capability instead. AGENTS.md is the portable convention most agents read, so a handoff that will cross tools belongs alongside it rather than in Claude-only configuration.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.