Debug Detective
$2.99OfficialUse when a failure's cause is unknown: a hypothesis-driven protocol from symptom to confirmed root cause, with evidence trail and regression test.
What you get
- ✓9-step procedure
- ✓1 ready-to-run code block
- ✓10-point quality checklist
- ✓8 pitfalls to avoid
- ✓Installs into 6 tools
- Version
- v1 →
- Last updated
- today
- Length
- 7 min read
- Requires
- Best with a strong model (Claude Sonnet 4)
Works in: Claude Code, Codex, Cline, opencode, OpenClaw, Hermes · Handles multi-file projects
Preview
When to use
Use when something misbehaves and the cause is unknown: a failing test, wrong output, crash, hang, flaky behavior, performance cliff, or "it worked yesterday". Switch to this protocol especially after one unsuccessful fix attempt — a failed patch is the signal to stop patching and start investigating. Do not use when the cause is already known (that is a fix, not an investigation) or for feature requests dressed as bug reports.
Inputs to gather
- The exact symptom, verbatim: the full error message and stack trace (never a paraphrase), the wrong output versus the expected output, and why the expected output is expected.
- Reproduction status: the exact command or steps,
…
🔒 Buy once ($2.99) to unlock the full playbook, download it, and install it in every tool you use.