agentleFS
mrzlimrzli/skills/skills/conversation/warmup/SKILL.md

warmup

Skill0 starsChanged 6 days agoMITView on GitHub

Help the user get back into the current project after a break, with a very short summary of where the work stopped and a few small steps to rebuild context. Use when the user returns to a project after a day or more away.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. missionbuilt/loadout★ 0same name
    2. Varlas/claude-skills★ 0same name

    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 repository17 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    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:

    Read these, stopping once the picture is clear:

    line 11
  2. Very specific:

    If the current session already holds earlier work, such as when it was resumed or continued, pass --skip 0 to print the current session …

    line 16
  3. Distinctive:

    Prefer written records and the repository over the session when they disagree, and mention the mismatch.

    line 19
  4. Distinctive:

    Write at most five short lines: what the user was doing, where it stopped, and what is waiting on them.

    line 23
  5. Very specific:

    If the project has had activity in the last four weeks but its latest retrospective record from improve-workflow is older than that, or …

    line 27
  6. Distinctive:

    Stop after the steps. Wait for the user to start.

    line 29

≈ “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

Get the user back into the current project quickly. Read only enough to know where the work stopped, not the whole history. Make no changes.

Find where the work stopped

Read these, stopping once the picture is clear:

  • The project's plan, queue, or progress files, such as .scratch/delivery/ queue files, the project's backlog, and any handoff document for this project.
  • The current branch, recent commits, unmerged branches, and uncommitted changes.
  • The end of the previous agent session in this project. Run last-session.sh from the project root. It prints the last messages of the most recent earlier Claude Code or Codex session, without tool output.
    • If the current session already holds earlier work, such as when it was resumed or continued, pass --skip 0 to print the current session instead.
    • With --skip 0, check that the printed session ends with this request, so it is the current session and not a fork of it. If it does not, raise --skip by one and check again, and tell the user when no printed session matches.

Prefer written records and the repository over the session when they disagree, and mention the mismatch.

Summarize and give steps

Write at most five short lines: what the user was doing, where it stopped, and what is waiting on them.

Then give three to five small steps in order, from easy to slightly harder. Each step names one concrete thing to look at or do, such as a commit, a file, a check to run, or a pending decision, and says what the user will know afterward. Use the project's real names and commands. Prefer steps that are part of the real work, such as reviewing the last commit or settling the next decision, over exercises. End with the step that starts the next piece of work.

If the project has had activity in the last four weeks but its latest retrospective record from improve-workflow is older than that, or there is none, add one line suggesting a retrospective with improve-workflow.

Stop after the steps. Wait for the user to start.

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.