agentleFS
nexus-a1nexus-a1/claude-skills/skills/bug-stub/SKILL.md

bug-stub

Skill1 starsChanged 28 days agoMITView on GitHub

Turn a free-text bug description into a paste-ready ticket stub — a short title and a one-paragraph summary — with no code discovery, no fix, and no commit. Use right after finding a bug, before any tracker ticket exists, to get text you can file yourself and then feed into no-ticket requirements gathering.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    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:

    If $ARGUMENTS is empty or too thin to extract these, use AskUserQuestion to ask for a one- or two-sentence description before continuing.

    line 41
  2. Very specific:

    Only invoke another skill on explicit confirmation.

    line 76

≈ “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

Bug Stub

Arguments: $ARGUMENTS

Purpose

You just found a bug. There's no ticket for it yet, and you don't want to stop and write one by hand. This command turns your free-text description into a title and summary you can paste into your own tracker (Jira, Linear, GitHub Issues, whatever you use) — nothing more.

This command does not investigate the bug. It performs no code discovery, proposes no fix, and makes no commit. If you want the bug actually investigated and fixed, use /troubleshoot instead — that command reads the codebase and reasons about root cause. This one only reshapes what you already told it.

Process

Step 1: Parse the description

Same parse shape as /troubleshoot's Phase 1 — this command borrows only that step, not anything past it. Extract from $ARGUMENTS (or ask if $ARGUMENTS is empty):

  • What: What component/endpoint/feature is broken?
  • Expected: What should happen?
  • Actual: What actually happens?
  • Repro: Steps to reproduce, if given.

If $ARGUMENTS is empty or too thin to extract these, use AskUserQuestion to ask for a one- or two-sentence description before continuing. Do not guess at missing details — an invented "expected" or "actual" would make the stub actively misleading to whoever files it.

Step 2: Generate the stub

Produce exactly two pieces of text:

  • Title — 80 characters or fewer, no markdown formatting (no backticks, no bold, no headers — plain text a tracker's title field will accept as-is). Describe the observable symptom, not the fix.
  • Summary — one paragraph covering what was expected, what actually happened, and how to reproduce it. No headers, no bullet list — a single paragraph, since it is meant to be pasted as a ticket description, not rendered as a mini-document.

Print both, clearly labeled:

Title:
{title}

Summary:
{summary}

Step 3: Offer the handoff (output chaining)

Offer — do not auto-run:

Stub ready. Want me to start requirements gathering from it? → /create-requirements --no-ticket Otherwise, copy the title/summary above into your tracker.

Only invoke another skill on explicit confirmation. If confirmed, print a handoff banner naming the exact next command and stop (same shape as /meeting's /epic handoff — never invoke the next skill directly):

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Handing off to /create-requirements
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Title:   {title}
Summary: {summary}

Next: run /create-requirements --no-ticket "{title}: {summary}"
No ticket number needed yet — a provisional draft session is created, and
you can reconcile it with a real ticket later once one exists.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

The stub text is directly usable as that command's input — no reformatting needed between what this command prints and what --no-ticket expects as a feature description.

Then stop. Do not run any further steps — /create-requirements takes over when the user runs it.

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.