agentleFS
philippmossierphilippmossier/cofabriq/skills/ticket/SKILL.md

ticket

Skill0 starsChanged 10 days agoMITView on GitHub

Turn a fuzzy note about a bug or a small feature into a GitHub issue cofabriq can take (`cofabriq intake`, `cofabriq run`). Use when the user types /ticket with a rough sentence, or asks to write up a bug or feature as a ticket.

23 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use orchestrate, retro and 7 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.
  • varun29ankuSSkill

    orchestrate

    This skill should be used when the user asks to 'orchestrate a task', 'break down work into parallel agents', 'coordinate subtasks', 'run agents in parallel', or mentions 'multi-agent'. Decomposes complex tasks into tracked subtasks, dispatches parallel subagents, and coordinates until completion.

    Also in 4 projects that use this

  • AWSSkill

    retro

    Conduct a retrospective on a coding session.

    Also in 3 projects that use this

  • SentrySkill

    spec

    Produce a verified spec — problem, desired outcome, and acceptance criteria — before building. Use when given a GitHub/Linear issue link, when starting from a fuzzy in-conversation idea, or asked to "figure out what to do" / "scope this" / "what should I build" — whenever the *what* isn't yet pinned down. Grills the ambiguities, checks the premise against the code, and produces acceptance criteria you sign off on — then hands to design-first.

    Also in 3 projects that use this

  • AWSSkill

    ship

    Finalize dev work on an issue branch — update docs, push, and open a PR.

    Also in 3 projects that use this

  • MicrosoftSkill

    handoff

    Compact the current conversation into a handoff document for another agent to pick up.

    Also in 5 projects that use this

  • GoogleSkill

    prototype

    Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 4 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 20 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the code just enough to name things the way the code does (the component, the function, the visible label).

    line 18
  2. Very specific:

    If the repository has cofabriq's MCP tools, use them as context: decisions (an ADR or rule the change would touch: say so in the ticket), …

    line 20
  3. Distinctive:

    Ask only what the ticket needs and the note doesn't say.

    line 23
  4. Distinctive:

    No steps, versions, error messages or expected behaviour the user didn't give or the code doesn't show.

    line 30
  5. Distinctive:

    Keep the user's own words where they carry the meaning.

    line 33
  6. Very specific:

    One slice: If it needs more than a few places in the code, split it into tickets and link them (Blocked by #N, or GitHub's own "blocked …

    line 41
  7. Very specific:

    No implementation plan: cofabriq's refine writes that from the code.

    line 49
  8. Very specific:

    Use the repository's own template if it has one (.github/ISSUE_TEMPLATE/).

    line 53

“Common” is measured: kept by at least 6 in 10 of the 20 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

/ticket: a fuzzy note in, a ticket cofabriq can take out

The user gives a rough note ("shift enter sends the message, should be a new line"). You turn it into one GitHub issue that cofabriq's readiness check will accept, without inventing anything. You draft; the user files.

1. Understand, then ask (one question at a time)

  • Bug or feature? A bug: something that worked or should work is broken. A feature: something new. Neither (a question, docs, a chore): say so and stop; cofabriq doesn't take those.
  • Read the code just enough to name things the way the code does (the component, the function, the visible label). Don't plan the fix.
  • If the repository has cofabriq's MCP tools, use them as context: decisions (an ADR or rule the change would touch: say so in the ticket), search (docs about this area), impact (likely components).
  • Ask only what the ticket needs and the note doesn't say. For a bug: steps, expected, actual, where (browser, OS, version). For a feature: what should happen, for whom, what must stay as it is. Ask one question at a time and stop asking once the ticket is checkable.

2. Never invent

  • No steps, versions, error messages or expected behaviour the user didn't give or the code doesn't show. A field you can't fill becomes an open question in the ticket ("Which browser?"), never a plausible guess.
  • Keep the user's own words where they carry the meaning.

3. Write it so cofabriq's readiness check passes

The check (cofabriq run, docs/intake.md) sends a spec back when it has no criterion a test can check, names files that don't exist, or is too big for one PR. So:

  • One slice. If it needs more than a few places in the code, split it into tickets and link them (Blocked by #N, or GitHub's own "blocked by").
  • Checkable behaviour. For a feature, 2-6 criteria, each a concrete situation and the exact expected result ("detectLang(["en-US", "de"]) returns "de""), plus what must stay as it is.
  • Name the interface a test will call when the feature adds one (a new return kind, a new function): the acceptance test is written against it and frozen.
  • No implementation plan: cofabriq's refine writes that from the code.

4. The ticket

Use the repository's own template if it has one (.github/ISSUE_TEMPLATE/). Otherwise:

Bug:

<one sentence: what is wrong>

**Steps**
1. ...

**Expected:** ...
**Actual:** ...
**Where:** <browser, OS, version, if known>

**Open questions**
- <anything only the user can answer, or none>

<details><summary>Original note</summary>

<the user's note, verbatim>
</details>

Feature:

<one sentence: what is wanted and why>

**What**
- <criterion: situation -> exact result>
- <what must stay as it is>

**Interface** (only if a test must call something new): ...

**Open questions**
- ...

<details><summary>Original note</summary>

<the user's note, verbatim>
</details>

The original note stays in every ticket: when a spec comes out wrong later, it shows whether the ticket or the refine lost something.

5. Show, then file

Show the title and body. File only after the user says so:

gh issue create --repo <owner/name> --title "<title>" --body-file <file>

Then tell them the next step: cofabriq intake <n> (order and readiness) or cofabriq run <n> --type bug|feature.

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.