agentleFS
alabenkhlifaalabenkhlifa/dayjob/skills/start-ticket/SKILL.md

start-ticket

Skill1 starsChanged 30 days agoMITView on GitHub

Use when the user starts work on a ticket, issue, story, or feature request (a ticket id like PROJ-123, a GitHub issue, or pasted requirements) and wants it understood, planned, and implemented.

8 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

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

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 8

    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:

    If nothing is unclear, skip the questions: put your one-line understanding at the top of the plan in step 4 and ask for a single yes.

    line 28
  2. Distinctive:

    Wait for a yes (one yes covers understanding and plan when there were no questions), unless the user pre-approved in step 2.

    line 44
  3. Distinctive:

    Only what the ticket asks: no extra features, no drive-by refactors.

    line 49
  4. Distinctive:

    No dead code, no unused imports.

    line 51
  5. Distinctive:

    Add or update tests for the new behavior when the project has tests.

    line 52
  6. Distinctive:

    No file edits before the user confirms your understanding, unless they pre-approved.

    line 67
  7. Distinctive:

    Never claim tests pass without running them.

    line 68
  8. Distinctive:

    When a requirement is ambiguous, ask.

    line 69

“Common” is measured: kept by at least 6 in 10 of the 7 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

Start ticket

Turn a ticket into the smallest correct change that follows the codebase's patterns: understand → plan → confirm → build → verify.

1. Get the ticket

  • A ticket id like PROJ-123 and a board tool is available (Jira, Linear, ClickUp, Trello, Asana, as MCP tools) → fetch summary, description, acceptance criteria.
  • A GitHub issue number or URL → gh issue view <n>, as the gh account picked below.
  • Otherwise use the text the user gave. If there is none, ask for it.

gh account: the user may have several gh accounts, and the active one may not see this repo. Before the first gh call, list each account's access:

for u in $(gh auth status 2>&1 | sed -n 's/.*account \([^ ]*\).*/\1/p'); do echo "$u $(GH_TOKEN=$(gh auth token --user "$u") gh repo view --json viewerPermission -q .viewerPermission 2>/dev/null)"; done; true

Pick the first account with ADMIN, MAINTAIN or WRITE, else one with READ or TRIAGE. If none shows a permission, skip every gh call. Run every gh call as that account, GH_TOKEN=$(gh auth token --user <account>) gh ..., with 2>/dev/null. Never run gh auth switch.

2. Find the gaps — stop here first

List what is unclear: expected behavior, inputs and validation, edge cases, error handling, UI, environments. Ask only the questions whose answers change the implementation, most important first, max 5. Then state your understanding in 3–5 lines and wait.

Do not edit or create any file until the user confirms. Exception: the user already said the requirements are final and to go ahead. Then state your understanding in one line and continue.

If nothing is unclear, skip the questions: put your one-line understanding at the top of the plan in step 4 and ask for a single yes.

3. Read the codebase

Find similar features to copy, helpers and constants to reuse, tests covering this area, and per-environment config. Cite path:line for each.

4. Plan — confirm before code

Change: <path> — <what>
Create: <path> — <why> (prefer none)
Reuse:  <path:line> — <what>
Tests:  <what you will add and run>
Risks:  <what could break>

Wait for a yes (one yes covers understanding and plan when there were no questions), unless the user pre-approved in step 2.

5. Build

  • Match the surrounding code's style and patterns exactly.
  • Only what the ticket asks: no extra features, no drive-by refactors.
  • Reuse existing helpers; use named constants, not magic values.
  • No dead code, no unused imports.
  • Add or update tests for the new behavior when the project has tests.

6. Verify

Find the build and test commands (package.json, Makefile, README, CI config) and run them. Fix failures you caused. Report:

Done:    <one line>
Changed: <files>
Tests:   <command> — <passed/failed counts>
Open:    <assumptions or follow-ups>

Rules

  • No file edits before the user confirms your understanding, unless they pre-approved.
  • Never claim tests pass without running them.
  • When a requirement is ambiguous, ask; don't guess.

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.