agentleFS
Sign inStart free
Domenico-Tenace-Open-LabsDomenico-Tenace-Open-Labs/web-dev-agent/skills/systematic-debugging/SKILL.md

systematic-debugging

Skill3 starsChanged 13 days agoMITView on GitHub

Fixing bugs, errors, unexpected behavior, failing tests. Not for new code. Without guidance, agents default to: read stack trace, spot suspicious line, change, hope. Sometimes works. Often fixes symptom not cause, or adds new bug. This skill slows that reflex, follows experienced developer path. Can't reproduce reliably = guessing, not debugging. Find minimal triggering case. Intermittent: find what differs between passing/failing runs (input, timing, prior state). Vague report ("it breaks sometimes"): turn into reproducible case first, don't poke code randomly.

190 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use chinese-code-review, chinese-commit-conventions and 10 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.
  • Leodorareluctant259Skill

    chinese-code-review

    中文代码审查规范——在保持专业严谨的同时，用符合国内团队文化的方式给出有效反馈

    Also in 8 projects that use this

  • Leodorareluctant259Skill

    chinese-commit-conventions

    中文 Git 提交规范 — 适配国内团队的 commit message 规范和 changelog 自动化

    Also in 8 projects that use this

  • obraSkill

    diagnosing-superpowers

    Use when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.

    Also in 20 projects that use this

  • mjunaidcaSkill

    building-chatgpt-apps

    Guides creation of ChatGPT Apps with interactive widgets using OpenAI Apps SDK and MCP servers. Use when building ChatGPT custom apps with visual UI components, embedded widgets, or rich interactive experiences. Covers widget architecture, MCP server setup with FastMCP, response metadata, and Developer Mode configuration. NOT when building standard MCP servers without widgets (use building-mcp-servers skill instead).

    Also in 4 projects that use this

  • mjunaidcaSkill

    building-chat-interfaces

    Build AI chat interfaces with custom backends, authentication, and context injection. Use when integrating chat UI with AI agents, adding auth to chat, injecting user/page context, or implementing httpOnly cookie proxies. Covers ChatKitServer, useChatKit, and MCP auth patterns. NOT when building simple chatbots without persistence or custom agent integration.

    Also in 4 projects that use this

  • +2 more from Leodorareluctant259
  • +5 more from mjunaidca

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 8 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 177 edited versions we read keep these.

  • 8

    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. Distinctive:

    No fix without clear theory of why.

    line 33
  2. Very specific:

    Before fixing, ask: fix cause or hide symptom?

    line 37
  3. Distinctive:

    Check nearby cases not broken (similar edge cases, alternate paths).

    line 45
  4. Distinctive:

    Run test suite if exists. No test covering this bug?

    line 46
  5. Distinctive:

    Not bureaucracy — lets reviewer trust change without redoing investigation.

    line 56
  6. Distinctive:

    Don't trial-and-error code changes hoping error disappears.

    line 60
  7. Distinctive:

    Don't ignore warnings/side errors during investigation — sometimes clue to real cause.

    line 61
  8. Distinctive:

    Don't mark resolved if main test passes but side effects unchecked.

    line 62
  9. Distinctive:

    Don't refactor while hunting bug.

    line 63

“Common” is measured: kept by at least 6 in 10 of the 177 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

Instructions

When to use it

Fixing bugs, errors, unexpected behavior, failing tests. Not for new code.

Why it matters

Without guidance, agents default to: read stack trace, spot suspicious line, change, hope. Sometimes works. Often fixes symptom not cause, or adds new bug. This skill slows that reflex, follows experienced developer path.

The process

1. Reproduce the problem before touching anything

Can't reproduce reliably = guessing, not debugging. Find minimal triggering case. Intermittent: find what differs between passing/failing runs (input, timing, prior state).

Vague report ("it breaks sometimes"): turn into reproducible case first, don't poke code randomly.

2. Isolate the cause before proposing a fix

Reproduced? Narrow it:

  • Bisection: suspect recent change? Check git history, find which commit introduced bug.
  • Targeted logging: logs/breakpoints at key flow points, not everywhere. Find where state diverges from expected.
  • Shrink the problem: bug in complex flow? Isolate in simpler scenario (standalone script, minimal unit test).

No fix without clear theory of why. "I changed this line and error went away" = luck, not explanation.

3. Tell the symptom apart from the cause

Before fixing, ask: fix cause or hide symptom? Classic: null crashes downstream. Adding if (x !== null) removes crash, but if x should never be null there, real bug is upstream where x produced.

Real cause too risky to fix now? Temporary mitigation OK, but flag clearly (code comment or PR note), don't pass off as final solution.

4. Verify the fix, not just the case you reproduced

After fix:

  • Confirm original case works.
  • Check nearby cases not broken (similar edge cases, alternate paths).
  • Run test suite if exists. No test covering this bug? Add one — best prevention against regression.

5. Explain what happened

Presenting fix, include:

  • Expected vs observed behavior.
  • Actual root cause (not just "I changed X").
  • Why fix addresses cause, not symptom.
  • Remaining risks or uncovered cases.

Not bureaucracy — lets reviewer trust change without redoing investigation.

Things to avoid

  • Don't trial-and-error code changes hoping error disappears.
  • Don't ignore warnings/side errors during investigation — sometimes clue to real cause.
  • Don't mark resolved if main test passes but side effects unchecked.
  • Don't refactor while hunting bug. Isolate + fix first, refactor separately if needed.

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