agentleFS
Sign inStart free
nadavnavehnadavnaveh/skills/skills/bug-triage/SKILL.md

bug-triage

Skill2 starsChanged 25 days agoMITView on GitHub

Checks a list of bug findings and keeps only the real ones. It reproduces each one, has a second agent try to disprove it, merges findings with one cause, writes what a user sees when it breaks, and starts each fix with a failing test. Fixes "the agent found ten bugs, and most of them were not real." Use when a review, an audit or an agent hands you a list of findings, before you fix any of them, or when several reports look alike. Also on /bug-triage.

29 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use ao-desktop-dev and pr-description.

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.
  • agentlab-inSkill

    ao-desktop-dev

    Launch, restart, or troubleshoot the real AO Electron desktop app from this repository; run a checkout against isolated or real local AO data; combine PR branches for local UI review; and diagnose stale Electron processes, port conflicts, or preload bridge mismatches. Use whenever asked to run, open, show, or visually verify AO frontend changes in the actual desktop app rather than ao preview, dev:web, or mock data.

    Also in 5 projects that use this

  • MicrosoftSkill

    pr-description

    Draft Nacelle PR titles, pull request descriptions, commit titles, commit descriptions, and squash commit messages. Use when asked for PR or commit wording, a branch summary, or a PR description that can be used in a commit. Omit validation sections from commit bodies and remove filler.

    Also in 4 projects that use this

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 27 edited versions we read keep these.

  • 5

    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:

    No repro, no bug. A finding you cannot make happen is a guess.

    line 11
  2. Distinctive:

    Every real bug gets a test that fails before the fix and passes after.

    line 15
  3. Distinctive:

    Write the smallest repro for each: a script, a command or steps to click through.

    line 19
  4. Distinctive:

    Never fix a finding that was not reproduced.

    line 27
  5. Distinctive:

    No finding ends as "looks fine".

    line 30

“Common” is measured: kept by at least 6 in 10 of the 27 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

Bug triage

The agent found ten bugs. Some were guesses, and three were the same bug. A finding reads like a fact, so it gets fixed like one. This skill checks every finding first and keeps only the bugs that survive.

The principles

  1. No repro, no bug. A finding you cannot make happen is a guess.
  2. A second agent argues it is not a bug. What survives is real.
  3. One cause is one bug, however many reports point at it.
  4. Harm sets the order: what a user sees, not what the code does.
  5. Every real bug gets a test that fails before the fix and passes after.

Steps

  1. Number every finding, with the file it points at and the claim in one line.
  2. Write the smallest repro for each: a script, a command or steps to click through. Run it. Mark it reproduced, with the output, or not reproduced, with what happened instead. Drop the second kind, or send it back to ask for steps.
  3. For each reproduced finding, start a subagent as the skeptic. Give it the finding, the repro and the output, and one job: argue it is not a bug (expected behavior, bad input, a path no user reaches). If it proves the case, mark it not a bug and drop it.
  4. Fold findings with one cause (the same null user, the same missing check) into the first one, and keep the count: "bug 1 (x3)".
  5. For each bug left, write three steps or fewer to make it happen, what the user sees (an error page, a wrong total) and a severity from that harm: blocks users, gives wrong results, or annoys.
  6. Show the table: each finding as real, not a bug, merged or not reproduced, with the reason. Ask "Fix these, in this order?" and start only after a yes.
  7. For each bug, in severity order, write a test named for it (test_bug_2). Run it and show it fail. Fix the bug, show the same test pass, then run the full test suite.

Rules

  • Never fix a finding that was not reproduced.
  • The skeptic gets the evidence, not your opinion.
  • A test that passes before the fix does not test the bug.
  • No finding ends as "looks fine". It is real, not a bug, merged or not reproduced.

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