agentleFS
Sign inStart free
ChianWChianW/C31/skills/core/C31-workflow-bug-reproduction/SKILL.md

C31-workflow-bug-reproduction

Skill1 starsChanged 3 months agoMITView on GitHub

Output language: Respond automatically in the user's conversation language. Systematic bug reproduction and isolation. Read-only — never writes code. User provides: - Bug symptom (what's wrong) - Environment where it occurs (OS, versions, config) - Any existing reproduction steps (even partial) - Whether it reproduces consistently or intermittently

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 11 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈11

    Distinctive · estimate

    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:

    Read-only workflow agent. Systematically reproduces bugs: hypothesize causes → construct minimal reproduction → verify.

    line 4
  2. Distinctive:

    Use when a bug report is vague, when regression tests are missing, or when separating environment bugs from code bugs.

    line 5
  3. Distinctive:

    Test with defaults before custom config

    line 49
  4. Distinctive:

    If intermittent, run N times and log frequency

    line 54
  5. Distinctive:

    Only reproduces on specific env?

    line 58
  6. Distinctive:

    If bug is intermittent, report reproduction rate ("3/10 runs")

    line 76
  7. Distinctive:

    Always test on clean environment before declaring "reproduced"

    line 77
  8. Distinctive:

    If cannot reproduce after 3 attempts, document everything tried and stop

    line 78
  9. Distinctive:

    Never mutate production data during reproduction

    line 79
  10. Distinctive:

    If reproduction requires sensitive data, use anonymized equivalent

    line 80
  11. Distinctive:

    Keep output under 80 lines

    line 81

≈ “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

Multilingual Triggers

LanguageTrigger phrases
ENC31-workflow-bug-reproduction
ZH复现bug, bug复现
JAバグ再現, 再現手順

Output language: Respond automatically in the user's conversation language.

C31-workflow-bug-reproduction

Systematic bug reproduction and isolation. Read-only — never writes code.

When to Use

  • Bug report lacks clear reproduction steps
  • Need to determine if bug is environmental or code-level
  • Creating a regression test for a reported issue
  • Bug only appears in CI/production but not locally
  • User says "it doesn't work" without specifics

Input

User provides:

  • Bug symptom (what's wrong)
  • Environment where it occurs (OS, versions, config)
  • Any existing reproduction steps (even partial)
  • Whether it reproduces consistently or intermittently

Reproduction Flow

  1. Form hypotheses — List 3 likely causes ranked by probability:
    • Code bug (logic error, race condition)
    • Environment issue (wrong version, missing dependency, stale state)
    • Config issue (wrong flag, missing env var)
  2. Minimize surface area:
    • Strip unrelated code/config
    • Use minimal dataset
    • Disable caching, async, or concurrency if possible
    • Test with defaults before custom config
  3. Controlled experiments:
    • Change one variable at a time
    • Document each result
    • If intermittent, run N times and log frequency
  4. Verify reproduction:
    • Reproduces on clean environment? → code bug
    • Only reproduces on specific env? → environment/config bug
    • Doesn't reproduce at all? → missing condition or fixed already

Output Format

## Bug Reproduction: [brief description]
**Hypotheses tested**:
| # | Hypothesis | Result | Evidence |
**Minimal reproduction**:
[steps or code that reliably triggers the bug]
**Root cause category**: [code / environment / config / cannot reproduce]
**Additional conditions required**: [if any]
**Recommended next step**: [fix / deeper investigation / environment change]

Rules

  • If bug is intermittent, report reproduction rate ("3/10 runs")
  • Always test on clean environment before declaring "reproduced"
  • If cannot reproduce after 3 attempts, document everything tried and stop
  • Never mutate production data during reproduction
  • If reproduction requires sensitive data, use anonymized equivalent
  • Keep output under 80 lines

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