agentleFS
hongzz0618hongzz0618/ai-agent-workflows/skills/start-ticket/SKILL.md

start-ticket

Skill1 starsChanged 22 days agoMITView on GitHub

Analyze a ticket and relevant repository evidence, then propose a minimal solution and verification plan for approval before implementation.

10 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

5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 9 edited versions we read keep these.

  • 5

    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:

    Keep this workflow read-only.

    line 12
  2. Distinctive:

    Use only inspection actions that do not change state.

    line 14
  3. Distinctive:

    Follow applicable repository instructions and existing conventions.

    line 16
  4. Distinctive:

    Use concise B1-B2 English.

    line 22
  5. Distinctive:

    If evidence is insufficient, make the recommendation conditional and state what must be confirmed before implementation.

    line 43

“Common” is measured: kept by at least 6 in 10 of the 9 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

Prepare a ticket for implementation using the five steps below.

Working Rules

  • Keep this workflow read-only. Do not edit files, install dependencies, change external resources, or start implementation.
  • Use only inspection actions that do not change state. Plan tests here; do not run checks that may write files or change data.
  • Follow applicable repository instructions and existing conventions.
  • Investigate only files relevant to the ticket. Expand the scope only when a clear dependency or missing evidence requires it.
  • Label verified facts, assumptions, and open questions separately.
  • Support facts and recommendations with file paths, relevant locations, or other available evidence. Do not present guesses as findings.
  • Use concise B1-B2 English. Keep technical names accurate.

1. Requirements Analysis

  • Summarize the problem, expected behavior, scope, and constraints.
  • Define observable acceptance criteria from the ticket. Mark any proposed criteria that still need confirmation.
  • Identify missing details. Ask only questions that affect the solution; continue independent investigation while answers are pending.

2. Repository Investigation

  • Find the relevant entry points, current behavior, and direct dependencies.
  • Inspect nearby tests and conventions needed to understand the change.
  • Report key facts with evidence, plus assumptions and unresolved questions.

3. Solution Analysis

  • Recommend one solution based on the requirements and repository evidence.
  • Explain why it fits and briefly compare an alternative when useful.
  • Include risks, compatibility concerns, and important tradeoffs.
  • If evidence is insufficient, make the recommendation conditional and state what must be confirmed before implementation.

4. Minimal Implementation Plan

  • List the smallest ordered steps and likely files or components to change.
  • Connect each step to a requirement. Avoid unrelated cleanup or refactoring.
  • Describe planned changes only; do not implement them.

5. Verification Plan

  • Map acceptance criteria to specific tests or observable checks.
  • Include relevant regression and failure cases, plus any manual checks.
  • State prerequisites, expected results, and verification limits.
  • Clearly mark checks as planned; never claim unperformed checks passed.

Present a concise report in this five-step order. End by asking for explicit approval of the proposed plan, then stop. Do not begin implementation until the user approves it and any blocking questions are resolved.

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.