agentleFS
Sign inStart free
odytriceodytrice/metaskills/skills/issue-refine/SKILL.md

issue-refine

Skill1 starsChanged 14 days agoMITView on GitHub

Clarify an issue's scope and acceptance criteria for implementation. Use to clarify, scope, groom, refine, or break down an existing issue.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. hirokisakabe/issuekit★ 1same name

    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.

  • In this repository15 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    No file lists, task breakdowns, data-model/API designs, or implementation order.

    line 16
  2. Distinctive:

    Ask only scope/acceptance questions.

    line 18
  3. Distinctive:

    No refinement comments. Discuss in chat.

    line 19
  4. Very specific:

    After approval, stage in .tmp-refined-body.md (file-write tool), apply, re-read, report, delete the file.

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Refine Issue

Own the What; issue-plan owns the How. No branches, commits, or implementation edits.

Project facts from AGENTS.md: § Repositories (issue template), § Project Board, § Code Layout & Tech Stack (feasibility checks only), § Review Notes (critical flows for bugs). Missing section: name it and stop.

Explicit board: none is valid per issue-plan § Board Status Transitions: skip board operations and report refinement without a status transition. Missing board facts still stop.

Rules

  • No file lists, task breakdowns, data-model/API designs, or implementation order. Convert design discussion to constraints in ## Notes (e.g. "must not change the public API"); defer design to issue-plan.
  • Code reading stops at feasibility: existing behavior, PR-sized scope, invariant conflicts.
  • Ask only scope/acceptance questions; distinguish facts from assumptions. Stop for blocking answers.
  • No refinement comments. Discuss in chat; body edits require approval. Non-interactive workers return proposal and questions.

Workflow

  1. Load issue and § Repositories template; preserve template sections.

    gh issue view <number-or-url> --json number,title,body,labels,assignees,comments,state,url
  2. Feasibility check (three questions above).
  3. Assess intent, behavior (flow, inputs/outputs, permissions, user-visible failures), boundaries, testable completion, and incoming/outgoing dependencies. Conclude: ready, needs answers, split, or needs diagnosis (§ Bugs).
  4. Discuss using the Output Shape.
  5. Propose template headings or ## Summary, ## Context, ## Expected outcome (acceptance criteria), ## Notes (binding constraints). In template task sections, leave a one-line plan-ledger pointer.

  6. After approval, stage in .tmp-refined-body.md (file-write tool), apply, re-read, report, delete the file.

    gh issue edit <number-or-url> --body-file .tmp-refined-body.md
  7. Move to ready per issue-plan § Board Status Transitions; report verified status. Split parents stay put; refine/ready children individually.

Bugs

Bugs and criteria per issue-diagnose § Criteria.

  • The body carries no fix, including as a ## Notes constraint; the expected outcome states behavior, not mechanism.
  • No <!-- root-cause --> comment and a criterion applies from issue text or history: conclude needs diagnosis, name the criterion, leave the issue in backlog. Resume after issue-diagnose posts.
  • RCA present: its Proof becomes an acceptance criterion ("the RCA reproduction passes"); apply its Refinement Needed; surface needs-evidence items as Questions.

Splitting

Beyond one PR: propose child titles and one-line scopes in chat under the rewrite approval gate. Retain parent with goal, context, links, and acceptance criterion "all sub-issues resolved". Create/link children via issue-raise § Sub-Issue Linking; report parent and every child. Never dissolve into loose top-level issues.

Output Shape

**Current Understanding**
**Proposed Scope**
**Out Of Scope**
**Acceptance Criteria**
**Constraints** (for the plan to respect; not the design)
**Dependencies**
**Questions** (numbered)

Omit empty sections.

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.

Check your setup, free