Skip to content
Agentmart
Submit a tool

systematic-debugging

Use when a bug, test failure, or unexpected behavior needs evidence-based diagnosis before a fix.

Skill by minfengyuan

IndexedMIT★ 1
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/minfengyuan-systematic-debugging

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when a bug, test failure, or unexpected behavior needs evidence-based diagnosis before a fix.

Use cases: testing

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found
Runs as
Node.js package

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/systematic-debugging

What's inside

Skill name
systematic-debugging
Description
Use when a bug, test failure, or unexpected behavior needs evidence-based diagnosis before a fix.
License
MIT
Instructions
33 lines in SKILL.md · view on GitHub

Before installing

  • Ships executable scripts

Files (6)

  • SKILL.md
  • condition-based-waiting-example.ts
  • condition-based-waiting.md
  • defense-in-depth.md
  • find-polluter.sh
  • root-cause-tracing.md

SKILL.md (first lines)

# Systematic Debugging

Find enough evidence to distinguish the root cause from the symptom, then make the smallest justified fix and verify it. Scale the investigation to the failure; do not apply a fixed ceremony to every issue.

## Workflow

1. Capture the exact failure, environment, recent changes, and reliable reproduction if available.
2. Trace the affected data or control flow to the boundary where observed behavior diverges from expected behavior.
3. Form a specific hypothesis and test it with the smallest useful diagnostic or change. Keep competing hypotheses separate.
4. Add or update a regression test when the behavior is testable and the test provides durable value.
5. Apply the narrow fix, run the failing check, and run related regression checks justified by the change.
6. If the evidence disproves the hypothesis, revise it rather than stacking unrelated fixes. If repeated attempts expose a design problem, pause and report the architectural decision needed.

## Evidence standards

- Read the complete error and relevant stack trace; do not infer a cause from one symptom.
- Compare a failing path with a known-good path when one exists.
- Inspect configuration, inputs, state propagation, and boundary conditions that can explain the failure.
- Use focused instrumentation or one-off probes when they reduce uncertainty; remove temporary diagnostics unless they are intentionally part of the fix.
- Keep unrelated refactors out of the debugging change.

## Completion

Report the reproduction or diagnostic evidence, root cause, changed files, checks run, and any unresolved uncertainty. Use the tdd skill when the user requests test-first development or when a regression test is the clearest way to drive the fix.

## Further resources

- [Root-cause tracing](root-cause-tracing.md)
- [Condition-based waiting](condition-based-waiting.md) and its [complete example](condition-based-waiting-example.ts)
- [Defense-in-depth validation](defense-in-depth.md)
- [Test-pollution bisection script](find-polluter.sh)

From minfengyuan/harness-knowledge-base, MIT.

More from minfengyuan/harness-knowledge-base

Signals

Stars
1
Forks
0
Open issues
0
Last commit (this folder)
Sep 16, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from minfengyuan

Cite this page

systematic-debugging skill. Agent Mart. https://agentmart.dev/skills/minfengyuan-systematic-debugging (last updated 2026-10-24)