agentleFS
bendyworksbendyworks/claude-skills/skills/pause-work/SKILL.md

pause-work

Skill1 starsChanged 37 days agoMITView on GitHub

Pause an unfinished story so a fresh session can carry it on. Writes dated resume notes into the story's plan file (what is pushed and what is only local, decisions with their reasons, checks already run, tests failing on purpose, what is waiting on the developer) and ends by naming the resume-work skill. Use when the user says "let's continue this in a fresh session", "I'm running low on context", "hand this off to a new session", "pause this story", or invokes the pause-work skill. Not for a shipped story: that is the finished-issue-housekeeping skill.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    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:

    End on one line: start a fresh session and invoke the resume-work skill with the plan's name, written the way this session invokes skills …

    line 69

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Pause work

This skill writes what the next session cannot work out from the repository into the story's plan file, where the resume-work skill (bundled in this plugin) reads it. Little context is usually left, so do only these steps.

The notes go in the story's plan file, never in memory and never only in chat. The plan-issue skill (bundled in this plugin) keeps plans at .work/plans/<name>.md. Look there first, then at .claude/plans/<name>.md for a plan made before .work/ existed, and write to the file where it is. A project may keep them elsewhere. With no plan file, ask the developer where to write them.

Steps

  1. Read the state now; never write it from recall.

    git branch --show-current
    git status --short
    git log --oneline <remote>/<branch>..   # commits not on the remote
    git stash list
    gh pr list --head <branch> --state all --json number,title,state,isDraft

    <remote> is the project's remote, usually origin; never use @{u}, which may be unset or name the base branch. With no <remote>/<branch>, the branch is not on the remote: say so, and list its commits since the base branch.

  2. Bring the plan up to date. Tick each to-do whose work is finished and verified, and correct any sentence in the plan about the story's state that is no longer true. When the plan names a GitHub issue, then sync that issue's checklist: gh-issue-sync checklist <issue> --plan <plan file>.
  3. Append the notes to the end of the plan file under ## Resume notes (<today's date>), after any earlier notes. Keep the heading at ##, where the to-do list ends for the tools that read it, and write no checkboxes under it: the finished-issue-housekeeping skill (bundled in this plugin) counts every unchecked box in the file as unfinished work. Refer to a to-do by its number. The next session acts on the newest notes only, so repeat what is still open. The notes say:

    • Pushed and local: the branch, each commit not on the remote (hash and subject), each uncommitted path and why (the plan file too, where git tracks it), each stash made on this branch, and each pull request with its state.
    • Decisions: each one made this session, with its reason.
    • Checks already run: each fact learned by running something, with the command and its result.
    • Failing on purpose: each such test by name, and what it waits for.
    • Waiting on the developer: each question or approval, with its text or the path to it.
    • Next: the next to-do by its number, where in it the work stopped, and which item above it waits on, if any.
  4. Correct memory; add nothing to it. Where the project's memory has an entry about this story's position that is now wrong, rewrite it as one true line pointing at the plan file. The notes never go there.
  5. End on one line: start a fresh session and invoke the resume-work skill with the plan's name, written the way this session invokes skills (/bendyworks:resume-work <name> where the plugin is installed). Print no prompt to copy.

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.