agentleFS
arbelharbelh/perdure/skills/handoff/SKILL.md

handoff

Skill0 starsChanged 29 days agoMITView on GitHub

Generate (or freshness-check) the HANDOFF page — a

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

2 rules. 0 are common, 2 are distinctive, and 0 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run the deterministic generator (never hand-write this page):

    line 12
  2. Distinctive:

    Before a handoff, a long pause, or archiving a session

    line 75

“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 — derived orientation page

Run the deterministic generator (never hand-write this page):

(${CLAUDE_PLUGIN_ROOT} is the plugin directory, set by Claude Code. Without it, the same scripts sit in the plugin checkout's root scripts/ directory (a Codex install keeps the whole checkout), in scripts/ beside this SKILL.md (a skills-CLI install), or in your clone of github.com/arbelh/perdure.)

python3 "${CLAUDE_PLUGIN_ROOT}/scripts/handoff.py" .

With --check as the argument, run integrity-check mode instead — it regenerates the page in memory and compares it to the committed one, with or without git, so it catches a stale page, a hand-edited one, and one left with merge-conflict markers. The stamp names the moment of the newest record update, not the commit or the clock, so it is compared like any other line. One fragment is normalized before comparing and is therefore NOT covered: the advisory ⚠ staleness annotations, whose commit counts move on their own. A page in the 4.2.0 format, with its commit-and-minute stamp and old footer, still passes when its content is current. Everything else must match the records exactly:

python3 "${CLAUDE_PLUGIN_ROOT}/scripts/handoff.py" . --check

Then:

  1. Report the generator's summary line (N records, M withheld for reconciliation).
  2. If records were withheld, list them and offer to reconcile: those records contradict themselves or git, and the page deliberately carries none of their claims. Fix them per the convention's Correcting a record rules (correct headers in place with a dated note; paraphrase retractions, never re-quote), then regenerate.
  3. Show the user the page path (perdure/HANDOFF.md) and, on first generation in a repo, a one-line explanation: derived, regenerated, never edited; every status is a claim as of the newest record update the stamp names.

On a merge conflict in the page, never resolve it by hand: regenerate it from the merged records (the command above) and commit the result. Its second line says the same, so any tool that opens the file sees it.

Why the lint gate is non-negotiable

A naive derived summary strips the context that lets a reader catch a stale claim, then presents the stale claim with the summary's authority. The generator therefore withholds all state claims from any record the rot lint flags. Do not paste record content into the page by hand around that gate; reconcile the record instead.

What the page does NOT claim

A lint-clean record can still be semantically stale, and external systems (experiments, deployments, tickets) are never checked. The page says this in its own header and footer — leave those disclaimers intact.

When to regenerate

  • At workflow close (after the record goes terminal and lint-clean) — the close creates the page if the project has none, and the close-gate hook regenerates a generator-stamped page on its own (opt-out for both: handoff-regen-at-close: false)
  • Before a handoff, a long pause, or archiving a session
  • Whenever --check reports the page no longer matches the records

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.