agentleFS
ShipfoxHQShipfoxHQ/shipfox/.agents/skills/implement-issue/SKILL.md

implement-issue

Skill87 starsChanged 12 days agoMITView on GitHub

Implement a Linear issue when given a Linear issue ID or URL. Use when the user invokes /implement-issue or asks to implement, build, or complete a specific Linear issue from an issue key, issue ID, or Linear URL.

31 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use review-issue, create-issue and 5 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.
  • ravihammondSkill

    review-issue

    Review and repair one GitHub issue implementation in the current checkout.

    Also in 4 projects that use this

  • sahinSkill

    create-issue

    Also in 6 projects that use this

  • TermiX-officialSkill

    merge-pr

    Merge a GitHub PR via squash after /preparepr. Use when asked to merge a ready PR. Do not push to main or modify code. Ensure the PR ends in MERGED state and clean up worktrees after success.

    Also in 3 projects that use this

  • ShopifySkill

    triage-issue

    Triage a GitHub issue — classify priority (P0/P1/P2), search for duplicates, and apply labels.

    Also in 3 projects that use this

  • SentrySkill

    create-pr

    Create pull requests following Sentry conventions. Use when opening PRs, writing PR descriptions, or preparing changes for review. Follows Sentry's code review guidelines.

    Also in 5 projects that use this

  • OpenAISkill

    writing-skills

    Use when creating new skills, editing existing skills, or verifying skills work before deployment

    Also in 3 projects that use this

The file

Show its rules and text

We split this skill into the rules it gives an agent

15 rules. 0 are common, 15 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 30 edited versions we read keep these.

  • 15

    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:

    Before starting implementation, gather the full Linear context for the requested issue:

    line 14
  2. Distinctive:

    Treat Linear as the source of truth for product intent, requirements, acceptance criteria, state, priority, ownership, and project context.

    line 23
  3. Distinctive:

    If any fetched Linear issue links to GitHub pull requests, fetch each pull request before implementation:

    line 27
  4. Distinctive:

    Read the pull request title, description, author, branch, review state, checks, and merge status.

    line 29
  5. Distinctive:

    Avoid duplicating work from merged or active pull requests unless the user explicitly asks for it.

    line 33
  6. Distinctive:

    Before making code changes:

    line 37
  7. Distinctive:

    Create or update a concise implementation plan from the gathered context.

    line 42
  8. Distinctive:

    If assigning to the current user is ambiguous, identify the current user through the available Linear tool context.

    line 44
  9. Distinctive:

    Prefer the smallest implementation that satisfies the Linear requirements and project context.

    line 51
  10. Distinctive:

    Run focused validation first, then broader validation when the change touches shared behavior.

    line 53
  11. Distinctive:

    Keep the Linear context in mind throughout, especially related issues and project-wide constraints.

    line 54
  12. Distinctive:

    If requirements conflict, are ambiguous, or force trade-offs, pause before implementation and offer the user the realistic options with the …

    line 56
  13. Distinctive:

    When implementation is complete:

    line 60
  14. Distinctive:

    Call out any remaining risks, skipped checks, or follow-up work.

    line 64
  15. Distinctive:

    Do not mark the Linear issue done unless the user explicitly asks for that status change.

    line 65

“Common” is measured: kept by at least 6 in 10 of the 30 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

Implement an issue

Required input

Require the user to provide a Linear issue ID, issue key, or Linear issue URL before doing anything else. If the user did not provide one, ask for it and stop until they answer.

Context gathering

Before starting implementation, gather the full Linear context for the requested issue:

  1. Fetch the issue itself and its comments.
  2. Fetch all associated issues, including parent, child, related, blocked-by, blocking, duplicate, and any other linked Linear relationships available through the Linear tools.
  3. Fetch the project the issue belongs to, if any.
  4. Fetch all other issues in that project.
  5. Fetch every document linked from any fetched issue or project.
  6. Continue following newly discovered linked issues, projects, and documents until the useful context graph has been covered or there is a clear cycle/repetition.

Treat Linear as the source of truth for product intent, requirements, acceptance criteria, state, priority, ownership, and project context.

GitHub pull request context

If any fetched Linear issue links to GitHub pull requests, fetch each pull request before implementation:

  1. Read the pull request title, description, author, branch, review state, checks, and merge status.
  2. Fetch the diff for each pull request.
  3. Identify whether the pull request is merged, open, draft, closed unmerged, or blocked.
  4. Incorporate relevant already-implemented changes into the plan.
  5. Avoid duplicating work from merged or active pull requests unless the user explicitly asks for it.

Start work

Before making code changes:

  1. Set the requested Linear issue status to In Progress.
  2. Assign the issue to the current user.
  3. Confirm the local repository context and read AGENTS.md and the sources it routes to before editing.
  4. Create or update a concise implementation plan from the gathered context.

If assigning to the current user is ambiguous, identify the current user through the available Linear tool context. If that is not available, ask the user which Linear assignee to use before changing the assignee.

Implementation workflow

Implement the issue end to end using the repository's existing conventions.

  1. Inspect the relevant code before editing.
  2. Prefer the smallest implementation that satisfies the Linear requirements and project context.
  3. Include tests when the behavior, contract, or bug risk warrants them.
  4. Run focused validation first, then broader validation when the change touches shared behavior.
  5. Keep the Linear context in mind throughout, especially related issues and project-wide constraints.

If requirements conflict, are ambiguous, or force trade-offs, pause before implementation and offer the user the realistic options with the consequences of each. Continue only after the user chooses a direction.

Completion

When implementation is complete:

  1. Report the files changed and the behavior implemented.
  2. Report the validation commands run and their results.
  3. Call out any remaining risks, skipped checks, or follow-up work.
  4. Do not mark the Linear issue done unless the user explicitly asks for that status change.

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.