Bug reproduction brief

Skill skyestrela/ai-agent-skill-preview/bug-reproduction-brief

Free evidence-first engineering skills for Claude Code, Codex, Hermes, Copilot and the skills CLI

Install

npx -y skills add skyestrela/ai-agent-skill-preview --skill bug-reproduction-brief

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

15 days oldThe repository was created 15 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Turn a vague bug report into a minimal, evidence-backed reproduction before proposing a fix.

SKILL.md

2.5 KB, 482 tokens by cl100k_base, as published. Nobody here has run it

Bug Reproduction Brief

Use this skill when a bug report is incomplete, intermittent, environment-specific, or mixed with an assumed cause.

1. Record the observed failure

Capture the exact error, incorrect output, timestamp, affected route/command and the smallest known input. Preserve logs without secrets or personal data. Label second-hand descriptions as unverified.

2. Identify the environment

Record only facts you can inspect: repository and commit, runtime version, operating system/container, dependency lockfile, relevant feature flags and whether the target is local, test or production. Never guess credentials or production configuration.

3. Write expected versus actual behaviour

Use two explicit statements:

Expected: [observable result]
Actual:   [observable result, including status/error]

Do not use the suspected cause as either statement.

4. Reduce the reproduction

Start from the reported path, then remove unrelated data, services and steps one at a time. Keep the smallest fixture that still fails. If it stops failing, restore the last removed condition and record it.

5. Prove repeatability

Run the minimal reproduction at least twice where safe. Record commands and outputs. If the failure is intermittent, report the observed frequency and duration instead of calling it deterministic.

Output

# Bug Reproduction Brief
- Target and commit:
- Environment:
- Report provenance: first-hand / second-hand; verified / unverified at intake
- Expected:
- Actual:
- Minimal steps:
- Minimal fixture:
- Reproduced: yes / no / intermittent
- Evidence:
- Unknowns:
- Safe next hypothesis to test:

Boundaries

Do not change production data merely to reproduce a bug.
Do not publish secrets, customer records or private source.
During this reproduction-only workflow, do not state a root cause or causal explanation, even when one line appears suspicious. Put only a testable candidate under Safe next hypothesis to test.
Stop after a verified reproduction; diagnosis and repair are separate workflows.

More evidence-first workflows

Free source and updates: https://github.com/skyestrela/ai-agent-skill-preview

Complete engineering pack: https://ai-agent-skills-pack.vercel.app/?utm_source=free-skill&utm_medium=github&utm_campaign=bug-reproduction-brief