agentleFS
drinkspillerdrinkspiller/skills/handoff/SKILL.md

handoff

Skill1 starsChanged 6 days agoUnlicenseView on GitHub

Compact the current conversation into a handoff document for another agent to pick up and reply with a copy-pastable prompt containing the full handoff path. Can accept arguments describing what the next session will focus on to tailor the document.

191 other projects use a version of this, 84 of them edited.

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

9 rules. 0 are common, 7 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 172 edited versions we read keep these.

  • 7

    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
  1. Distinctive:

    Write a handoff document summarising the current conversation so a fresh agent can continue the work.

    line 10
  2. Distinctive:

    Read Before Write: Read the file (or verify it is empty/accessible) before writing to it to ensure permissions.

    line 17
  3. Distinctive:

    Separate Signal from Noise: Before writing, mentally separate actionable context (decisions, requirements, designs, pending work) from …

    line 19
  4. Very specific:

    Project plans / task trackers: Check for any structured plan that was active during the conversation (e.g.

    line 26
  5. Distinctive:

    Open issues: If issue-tracker items (e.g.

    line 33
  6. Distinctive:

    Open PRs / CLs: Check for any pull requests or changelists that were created or modified but not yet merged/submitted.

    line 37
  7. Distinctive:

    Avoid Duplication: Do not duplicate content already captured in other artifacts (PRDs, plans, task trackers, issues, commits, diffs).

    line 67
  8. Distinctive:

    Tailor to Next Steps: If the user passed arguments or specified what the next session will focus on, tailor the handoff document …

    line 71
  9. Very specific:

    Output Path & Copy-Pastable Next-Agent Prompt: Always inform the user of the full path where the handoff document was saved, and always …

    line 74

“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 Skill

<!-- Originally based on https://github.com/mattpocock/skills/tree/main/skills/productivity/handoff -->

Write a handoff document summarising the current conversation so a fresh agent can continue the work.

Instructions

  1. Generate Handoff File Path: Generate a temporary file path using mktemp -t handoff-XXXXXX.md.
  2. Read Before Write: Read the file (or verify it is empty/accessible) before writing to it to ensure permissions.
  3. Separate Signal from Noise: Before writing, mentally separate actionable context (decisions, requirements, designs, pending work) from noise (debug logs, raw tool output, abandoned ideas, transient snippets). Only the signal goes into the handoff document.
  4. Gather Remaining Work: Actively search for outstanding work items. This is critical — handoffs that omit remaining work cause the next session to miss important context.
    • Project plans / task trackers: Check for any structured plan that was active during the conversation (e.g. a project management framework like Conductor, Linear, Jira, or GitHub Projects; a plan.md or TODO list; a design doc action-items section; or any markdown file with task checkboxes). List all unchecked (- [ ]) items, grouped by phase or category. Reference the plan file path rather than copying full task descriptions.
    • Open issues: If issue-tracker items (e.g. Buganizer, Jira, GitHub Issues, Linear) were referenced or created during the conversation, fetch each one and include its ID, title, status, and dependency chain. Note which issues block others.
    • Open PRs / CLs: Check for any pull requests or changelists that were created or modified but not yet merged/submitted. Include their review status.
    • Conversation context: Scan the conversation for any explicit follow-up items the user mentioned or acknowledged (e.g. "we'll do X later", "TODO: Y").
  5. Write with a Two-Pass Structure: First, write a 2-3 sentence anchor summary that captures the core state of the project right now. Then expand around that anchor with the following sections:
    • Conversation lineage — include the current session/conversation ID and any relevant related conversation IDs (e.g., from sub-sessions, parent sessions, or parallel tracks) to allow tracing the history if the summary is insufficient.
    • What has been completed (with CL/PR numbers and merge status).
    • Remaining work — the items gathered in step 4, organised by priority. Include dependency chains where they exist.
    • Known issues / risks — blockers, flaky infrastructure, environmental gotchas, or unresolved ambiguities that are not yet tracked as formal issues but would trip up the next session.
    • Problem-solving history — document what was tried and failed, dead ends, and rejected approaches so the next session does not repeat them. Include error messages or symptoms where relevant.
    • Key technical concepts — list technologies, coding conventions, architectural patterns, and domain terms discussed that a fresh agent would need to calibrate on.
    • Key file references (table of paths and URLs), with a brief note on why each file matters or what changed in it.
    • Next steps (place this last in the document for maximum visibility): recommended execution order and suggested skills to be used by the next session.
  6. Avoid Duplication: Do not duplicate content already captured in other artifacts (PRDs, plans, task trackers, issues, commits, diffs). Reference them by path or URL instead. For plan files with checkboxes, list only unchecked items — do not repeat checked ones.
  7. Tailor to Next Steps: If the user passed arguments or specified what the next session will focus on, tailor the handoff document accordingly to emphasize relevant context.
  8. Output Path & Copy-Pastable Next-Agent Prompt: Always inform the user of the full path where the handoff document was saved, and always include a copy-pastable prompt (in a fenced text code block) for the user to paste into the next agent session. The prompt must include the full path to the handoff document and a brief directive on what to pick up next. For example:

    Read /tmp/handoff-XXXXXX.md to load context from the previous session, then continue with the Next Steps section.

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.