agentleFS
shuvamkshuvamk/openbench/.claude/skills/file-issue/SKILL.md

file-issue

Skill2 starsChanged 5 months agoApache-2.0View on GitHub

File a finding or task into the SDLC queue (shuvamk/openbench) as a GitHub issue labeled status:ready, with fingerprint-based dedupe so re-filing the same finding just bumps the existing issue. Trigger when the user says "file an issue", "open a ticket for this", "/file-issue", when a worker discovers out-of-scope work mid-fix and must file a follow-up instead of expanding scope, or when any audit/review surfaces something worth queuing. Not for planner-authored milestone breakdowns (that's the planner agent) — this is for ad-hoc, single findings.

14 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use review-pr.

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.
  • DockerSkill

    review-pr

    Review one or more incoming Docker documentation pull requests as a maintainer. Independently validate technical claims, assess editorial fit and information architecture, choose a verdict, and draft exact inline or PR-wide feedback behind a confirmation gate. Use for requests such as "review PR 123", "is this PR correct?", "does this information belong here?", "validate this PR", or "help review backlog PRs". Do not use to maintain or fix a PR you own; use maintain-pr for that.

    Also in 4 projects that use this

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 13 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

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

    Run from the main checkout (absolute path or cd to repo root).

    line 12
  2. Distinctive:

    Don't use it for vague "improve X" wishes — a filable issue needs evidence and acceptance criteria a worker can turn into a failing test.

    line 21

“Common” is measured: kept by at least 6 in 10 of the 13 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

/file-issue — Queue a single finding

Wraps .claude/scripts/sdlc/file-issue.sh. Turns one finding into a well-formed status:ready issue that /pick-issue can later claim.

Run from the main checkout (absolute path or cd to repo root).

When to use it

  • A worker (in /pick-issue) finds scope creep — file a follow-up, keep the current fix tight.
  • A review/audit surfaces a concrete, actionable defect with file:line evidence.
  • The user points at something and says "make a ticket."

Don't use it for vague "improve X" wishes — a filable issue needs evidence and acceptance criteria a worker can turn into a failing test.

Invocation

.claude/scripts/sdlc/file-issue.sh \
  --title       "ERC misses floating output pins" \
  --priority    p1 \
  --area        agent-ops \
  --type        bug \
  --evidence    "packages/erc/src/rules.ts:42 — no driver-count check" \
  --evidence    "test/erc.test.ts — case not covered" \
  --impact      "Floating outputs pass ERC silently; user ships a broken schematic" \
  --suggestion  "Add a driver-count rule; 0 drivers on an OUTPUT net is a violation" \
  --verification "New failing test in erc.test.ts goes green" \
  --source      "manual"

Required flags

  • --title — imperative, specific.
  • --priority — p0 | p1 | p2 (bare labels, per .github/LABELS.md).
  • --area — one of: ir-schema, mcp-kicad, mcp-sim, mcp-firmware, registry, frontend, ai, collab-engine, agent-ops.
  • --type — one of: feature, bug, test, refactor, spike, infra, docs.
  • --evidence — repeatable; each is a path/file:line — what's wrong citation.

Recommended flags

  • --impact — why it matters (helps the worker sanity-check scope).
  • --suggestion — a direction, not gospel.
  • --verification — the acceptance criteria, phrased so a worker can write it as a failing test first. This is the most valuable field for the TDD contract.
  • --source — free-text provenance tag (default manual); part of the dedupe fingerprint.

Output

  • On success: the new issue URL. Labels applied: status:ready, <priority>, area:<area>, type:<type>, agent:planned.
  • DUPLICATE: <url> — an open issue with the same fingerprint already exists; the script posted a "re-detected" comment instead of filing a dupe. Not an error.

Notes

  • Dedupe is by a sha1 fingerprint of source + title + sorted evidence, embedded as <!-- fingerprint:... --> in the body and matched via GitHub search. Re-running the exact same call is safe.
  • The issue lands as status:ready and is immediately claimable by /pick-issue.
  • To file a follow-up from inside a worker run, still call this from the main checkout, capture the URL, and mention it in your PR / hand-off.

Sibling skills: /pick-issue (claim & work), /reap-stale (reaper). Pipeline docs: .claude/sdlc/README.md.

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.