crew-debugging

Files (9)

• SKILL.md
• condition-based-waiting.md
• defense-in-depth.md
• find-polluter.sh
• root-cause-tracing.md
• test-academic.md
• test-pressure-1.md
• test-pressure-2.md
• test-pressure-3.md

SKILL.md (first lines)

# Systematic debugging

**Adapted from `superpowers:systematic-debugging` by Jesse Vincent, MIT
licensed. The full copyright and permission notice is in
`plugin/crew/NOTICE.md`.** Upstream: <https://github.com/obra/superpowers>.
The method below - the Iron Law, the four phases, the red flags and the
rationalisation table - is upstream's and is reproduced with its wording
intact. What crew changed is **where Phase 1 gets its evidence**: this
repository already wrote down what breaks here, and a debugging method that
re-derives it from the diff is throwing that away.

## Prefer the upstream skill when it is installed

`superpowers:systematic-debugging` is the source of this method and is
maintained. When it is present, use it.

1. Invoke `superpowers:systematic-debugging` with the `Skill` tool.
2. If that invocation returns, you are on the upstream method. Come back here
   for **"Phase 1 in a crew repository"** below, which is the part upstream
   does not have, and follow everything else from upstream.
3. If the invocation errors, the skill is not available - use this file, which
   is a complete copy of the method and needs nothing else.

**Say in your report which path you took.** "Ran `superpowers:systematic-debugging`
plus crew's Phase 1 evidence sources" and "ran crew's inline copy" are
different provenance, and a reader who cannot tell them apart cannot tell
whether they are reading upstream's current wording or crew's snapshot of it.

**The honest limit on that check.** A failed `Skill` invocation proves the
skill is absent **from this session** - not from the machine. A plugin can be
installed and disabled, or enabled in another project, and this session sees
the same error in every case. So report "not available in this session", never
"not installed". Do not go looking in the plugin cache to settle it: that is
crew reimplementing the harness's own resolution, it goes stale the moment the
harness changes, and the answer it produces would not change what you do next.

**This is a preference between two present methods, not a survival path.**
Crew bundles the full copy, so nothing is lost when upstream is missing; the
routing only decides whose wording you read. Do not report an absent
`superpowers` as a degraded run.

## Overview

**Core principle:** ALWAYS find root cause before attempting fixes. Symptom
fixes are failure.

**Violating the letter of this process is violating the spirit of debugging.**

## The Iron Law

```
NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST
```

If you haven't completed Phase 1, you cannot propose fixes.

## When to Use

Use for ANY technical issue:
- Test failures
- Bugs in production
- Unexpected behavior
- Performance problems
- Build failures
- Integration issues

**Use this ESPECIALLY when:**
- Under time pressure (emergencies make guessing tempting)
- "Just one quick fix" seems obvious
- You've already tried multiple fixes
- Previous fix didn't work
- You don't fully understand the issue

**Don't skip when:**
- Issue seems simple (simple bugs have root causes too)
- You're in a hurry (rushing guarantees rework)
- Manager wants it fixed NOW (systematic is faster than thrashing)

## The Four Phases

You MUST complete each phase before proceeding to the next.

### Phase 1: Root Cause Investigation

**BEFORE attempting ANY fix:**

1. **Read Error Messages Carefully**
   - Don't skip past errors or warnings
   - They often contain the exact solution
   - Read stack traces completely
   - Note line numbers, file paths, error codes

2. **Reproduce Consistently**
   - Can you trigger it reliably?
   - What are the exact steps?
   - Does it happen every time?
   - If not reproducible -> gather more data, don't guess

3. **Check Recent Changes**
   - What changed that could cause this?
   - Git diff, recent commits
   - New dependencies, config changes
   - Environmental differences

4. **Gather Evidence in Multi-Component Systems**

   **WHEN system has multiple components (CI -> build -> signing, API ->
   service -> database):**

   **BEFORE proposing fixes, add diagnostic instrumentation:**
   ```
   For EACH component boundary:
     - Log what data enters component
     - Log what data exits component
     - Verify environment/config propagation
     - Check state at each layer

   Run once to gather evidence showing WHERE it breaks
   THEN analyze evidence to identify failing component

From mbadali25/useful-claude-add-ons, GPL-2.0.

More from mbadali25/useful-claude-add-ons