agentleFS
Alex5350Alex5350/arsenal/.agents/skills/work-tracking/SKILL.md

work-tracking

Skill0 starsChanged 16 days agoMITView on GitHub

Create, comment on, transition, link, and show work items in GitHub Issues, Jira, or Azure DevOps, and open the review pull request against one, so the tracker always matches reality. Use when starting any unit of work, at every loop phase transition, when opening a PR, and when asked to file, update, close, or check a ticket or issue.

  • Reads credentials

4 other projects use 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. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 0

    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:

    Only update states you can prove: show <id> is the proof.

    line 54

“Common” is measured: kept by at least 6 in 10 of the 4 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

Work tracking

Goal: the tracker is the heartbeat, not a diary. If work happened, the tracker says so, in the same session. Phase doc: docs/workflows/05-tracking.md.

Mechanics

One command, three backends (configured via .env, shape in docs/templates/work-item.example.env):

scripts/work-item.sh create   "Summary" --body "Context" [--type bug|task|story]
scripts/work-item.sh comment  <id> "What changed and the evidence"
scripts/work-item.sh transition <id> "<state>"
scripts/work-item.sh link     <id> <url>
scripts/work-item.sh pr       <id> "Title" [--body "..."] [--base <branch>]
scripts/work-item.sh show     <id>

pr opens the review pull request with the ID stamped in the title and the backend's closing reference in the body (Closes #<id>, the Jira key, or AB#<id>); it needs the gh CLI on every backend. show prints the item's current state on one line: run it before claiming a state.

Backend selection: WORK_BACKEND=gh|jira|ado in .env. Per-backend setup: docs/integrations/.

State mapping

Loop eventTracker action
question acceptedcreate item, state backlog
plan approvedtransition to ready, link the plan doc
execution startstransition to in progress; branch named <tracker>-<id>-<slug>
verification requestedpr opens the review PR (on GitHub it also labels the issue in-review; on Jira and Azure DevOps transition to in review yourself); comment the handoff note
gate verdictcomment evidence; the item stays open until the PR merges
PR mergedshow confirms done/closed (GitHub closes the issue from the PR body); transition <id> done if it did not
publishedcomment the release link; transition to released/closed

Adapt the state names to the backend's workflow once and record the mapping in docs/workflows/05-tracking.md of the consumer repo.

Rules

  • One unit of work, one item, one branch. No item, no branch.
  • IDs are stamped in branch names, commits, and PR titles.
  • Comments are short and factual: state changes, evidence links, verdicts. The tracker is an interface, not a transcript.
  • Only update states you can prove: show <id> is the proof. A tracker claiming "done" on a red gate manufactures confidence and is a worse failure than no tracking.

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.