agentleFS
javierbreajavierbrea/eslint-plugin-boundaries/.agents/skills/create-issue/SKILL.md

create-issue

Skill1k starsChanged 11 days agoMITView on GitHub

Create one or several GitHub issues of this repository (feature, task, bug, or the sub-issues of a tracking issue or an accepted RFC) with the structure, labels and bookkeeping the repository uses — labels, assignee, milestone, project board column and parent issue — showing a preview before anything is created. Use when asked to create, draft, file or break down issues.

We split this skill into the rules it gives an agent

5 rules. 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 49 edited versions we read keep these.

  • 1

    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. Very specific:

    Before drafting, search the open issues (gh issue list --state open --search "<words>") so that an existing issue is reused or linked …

    line 21
  2. Very specific:

    Check the labels, milestones and board fields on GitHub before using them (gh label list, gh api repos/{owner}/{repo}/milestones, gh …

    line 41
  3. Distinctive:

    Create the issues in dependency order (parents and blockers first, so that their numbers can be referenced), then do the bookkeeping of …

    line 54
  4. Very specific:

    When a tracking issue gets new sub-issues, update its Plan section with their numbers.

    line 55
  5. Very specific:

    Prefer the GitHub MCP tools when they are available and fall back to gh.

    line 58

“Common” is measured: kept by at least 6 in 10 of the 49 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

Create issue

When to use this skill

Use it to file any issue in this repository, from a one-line task to the set of sub-issues of an RFC. The workflow it implements is docs/contributing-workflow.md (steps 2 to 4); this skill adds the structure, the exact labels and the commands. Write every text with the writing-style skill.

Inputs

Gather from the request and the conversation:

  • What each issue delivers and why. For sub-issues, the parent issue or the RFC they come from.
  • Milestone, when the work already has one. Otherwise propose one (see "Bookkeeping").
  • Assignee: the maintainer unless told otherwise.

Before drafting, search the open issues (gh issue list --state open --search "<words>") so that an existing issue is reused or linked instead of duplicated.

Structure

Title. An imperative sentence naming the outcome, with the package when it matters: Make the browser build of @boundaries/elements work without Node.js shims. The type goes in the labels, not in a prefix.

Body of a bug. The .github/ISSUE_TEMPLATE/bug_report.md template.

Body of any other issue, in this order, dropping a section that would be empty:

  1. ## Context: why the issue exists, in a paragraph. An issue born from an RFC opens with From [RFC #N: title](url) and does not repeat the RFC's reasoning; an issue born from another issue links it.
  2. ## Proposal (or ## Scope for a task): what changes, as a list or a short numbered plan, naming the files, packages and commands involved. What is explicitly out of scope goes at the end of this section.
  3. ## Acceptance criteria: a checklist that a reviewer can tick without asking the author. Each item is observable (a command that passes, a page that renders, a file that exists).
  4. ## Open questions: questions the implementer cannot answer alone, each with who decides it.
  5. ## Blocked by / ## Related: issue numbers with a few words each on why.

A tracking issue (the parent of sub-issues) replaces Proposal with ## Plan: the ordered list of its sub-issues, each as - [ ] #N so that GitHub shows the progress.

Bookkeeping

Check the labels, milestones and board fields on GitHub before using them (gh label list, gh api repos/{owner}/{repo}/milestones, gh project field-list 7 --owner javierbrea); the lists below are the ones in use and are not updated by this skill.

  • Labels. One type: label: type: feature for new behavior, type: bug for a defect, type: task for work with no new behavior; a sub-issue takes the type of its own work. type: feature request and type: support are for issues opened by users. Add the scope: labels of the files the work touches: scope: code (source under packages/*/src), scope: tests (test/, *.spec.*, *.test.*), scope: documentation (docs, website content, *.md), scope: CI-CD, scope: dependencies, scope: refactor. Add status: blocked when a Blocked by section exists. Match the strings exactly, including the space after the colon.
  • Milestone. Work shipped by a package release goes in the version milestone vX.Y.Z. Work that spans releases and comes from an RFC gets a milestone named after the feature, with the RFC in its description. Create a missing milestone with gh api repos/{owner}/{repo}/milestones -f title=... -f description=....
  • Project board. Every issue goes to the user project "JS Boundaries" (number 7) with Status set to Backlog: gh project item-add 7 --owner javierbrea --url <issue url> --format json --jq .id returns the item id, then gh project item-edit --project-id <project id> --id <item id> --field-id <status field id> --single-select-option-id <backlog option id>. The project id comes from gh project view 7 --owner javierbrea --format json --jq .id, and the Status field and Backlog option ids from gh project field-list 7 --owner javierbrea --format json; the commands need the project token scope (gh auth refresh -s project). Leave the board's Priority and Size fields to the maintainer.
  • Assignee. --assignee on creation.
  • Parent issue. A sub-issue is attached to its parent with the GraphQL mutation addSubIssue(input: { issueId, subIssueId }), using the node ids from gh issue view N --json id; gh issue create has no option for it.

Procedure

  1. Resolve the inputs and search for duplicates.
  2. Draft every issue: title, body, labels, milestone, assignee, parent.
  3. Show a preview of all of them and ask for explicit confirmation. Nothing is created before it.
  4. Create the issues in dependency order (parents and blockers first, so that their numbers can be referenced), then do the bookkeeping of each one.
  5. When a tracking issue gets new sub-issues, update its Plan section with their numbers.
  6. Report the URLs, the milestone and the board column.

Prefer the GitHub MCP tools when they are available and fall back to gh. Treat an authentication or authorization error as terminal: stop, report it, and include the drafted titles and bodies in the response so that they can be filed by hand.

Notes

  • One issue per deliverable that gets its own pull request. Research that only informs a decision is a comment on the tracking issue, not an issue, unless it is more than a day of work.
  • The issue is not the plan: steps, file maps and commands that only the implementer needs go to .agents/plans/ (write-plan skill), and the issue links the plan's branch when one exists.

Download now (ZIP)

  • Other versions58 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 58

    1. penpot/penpot★ 62ksame name
    2. videojs/video.js★ 40ksame name
    3. yugabyte/yugabyte-db★ 11ksame name
    4. antiwork/gumroad★ 10ksame name
    5. gitkraken/vscode-gitlens★ 10ksame name
    6. camunda/camunda★ 4.3ksame name
    7. robustmq/robustmq★ 1.8ksame name
    8. aws-powertools/powertools-lambda-typescript★ 1.8ksame name
    9. classroomio/classroomio★ 1.7ksame name
    10. tikv/pd★ 1.2ksame name

    48 more not listed

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • Used alongsideProjects with it also use native-nostalgia-stack, roadmap-review and 8 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.
    • frostneySkill

      native-nostalgia-stack

      Applies the user's FreePascal toolchain and its build, formatting, hook, and test contracts while leaving project-specific mechanics local. Use when scaffolding or working in a FreePascal project that follows this toolchain.

      Also in 3 projects that use this

    • frostneySkill

      roadmap-review

      Reviews a roadmap from fresh project evidence and produces a verified, throughput-anchored version plan, with execution gated on confirmation. Use when reviewing a roadmap, planning releases, or sequencing a backlog.

      Also in 3 projects that use this

    • GoogleSkill

      implement-issue

      Implement a Site Kit GitHub issue end-to-end from its implementation brief. Use when the user asks to implement, build, or work on a GitHub issue / feature by number (e.g. "implement issue #12345", "build the feature in issue 12345", "follow the implementation brief for #12345") in the google/site-kit-wp repo.

      Also in 6 projects that use this

    • MicrosoftSkill

      create-release

      Prepare BC-Bench release notes for review. Use when creating a release, screening merged PRs since the previous version tag for changes that may affect evaluation results, matching recent GitHub release notes, or reporting the agent harness and development tool versions shipped by a release.

      Also in 4 projects that use this

    • MicrosoftSkill

      update-pr

      Update the pull request for the current session. Use when the user wants to push new changes to an existing PR.

      Also in 3 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

    • +1 more from frostney
    • +1 more from microsoft
  • In this repository12 other files it gives its agents.

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

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.