Debugger
Systematic debugging method: 5-step root-cause analysis (capture, isolate, hypothesize, investigate, fix & verify) plus common bug-pattern reference. Use when errors, exceptions, test failures, or unexpected behavior appear. Loaded automatically by the debugger agent.From its SKILL.md
npx -y skills add claude-world/director-mode-lite --skill debuggerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
SKILL.md
3.6 KB, 811 tokens by cl100k_base, as published. Nobody here has run it
Debugger Skill
Director Mode Lite - Debugging Specialist
5-Step Root-Cause Method
1. CAPTURE → Collect the full error, reproduce, define expected vs actual
2. ISOLATE → Narrow to the smallest failing case
3. HYPOTHESIZE → List and rank likely causes
4. INVESTIGATE → Test hypotheses, inspect state
5. FIX & VERIFY → Apply the minimal fix, confirm, prevent recurrence
Step 1: Capture
- Collect the complete error message and stack trace
- Can reproduce the issue with clear steps
- Identify the input that triggered the error
- Know expected vs actual behavior
Step 2: Isolate
- Identify the failure location from the stack trace
- Which file(s) and function(s) are involved?
- When did it start? (
git bisect,git log) - Narrow down to the smallest reproducible case
Step 3: Hypothesize
- List possible causes based on evidence
- Rank hypotheses by likelihood
- Design a test to confirm or eliminate each
Step 4: Investigate
- Analyze error messages and logs carefully
- Add strategic debug logging; inspect variable state at key points
- Check for: null/undefined values, type mismatches, race conditions, resource exhaustion, external service failures
Common Bug Patterns
| Pattern | Signs | Common Fix |
|---|---|---|
| Null/Undefined | Cannot read property of undefined | Add null checks / optional chaining |
| Off-by-one | Loop runs one too many/few times | Check loop bounds |
| Race condition | Intermittent failures | Add synchronization |
| Type coercion | "1" + 1 = "11" | Explicit type conversion |
| Async issues | Promise { <pending> } | Await/handle promises |
By language and domain:
- JavaScript/TypeScript:
undefined is not a function(binding),Cannot read property of null(optional chaining), promise rejections (async/await handling), type errors (strict mode) - Python:
AttributeError(init/typos),TypeError(type validation),ImportError(paths, circular imports),KeyError(dict defaults) - Database: connection timeouts (pool exhaustion), constraint violations (validation, foreign keys), deadlocks (transaction ordering)
- API/Network: 4xx (request validation, auth), 5xx (server-side, resource limits), timeouts (network, long-running queries)
Investigation Tools
# Search for error message
grep -r "error message" src/
# Find recent changes
git log --oneline -20
git diff HEAD~5
# Check specific function
grep -r "functionName" src/
Step 5: Fix & Verify
Apply the solution:
- Make the minimal change; don't refactor while fixing; one fix per commit
Confirm the fix:
- Original issue is resolved
- No new issues introduced
- Tests pass; add a test to prevent recurrence
- Manual verification done
Output Format
## Debug Report
### Issue
[Description of the bug]
### Reproduction Steps
1. Step one
2. Step two
3. Observe error
### Root Cause
[Explanation of why this happens]
### Location
- **File**: `src/utils/parser.ts`
- **Line**: 45-52
- **Function**: `parseInput()`
### Fix Applied
[Description of the fix]
### Verification
- [x] Issue resolved
- [x] Tests pass
- [x] No regression
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most research analysis skills give in 811 tokens
Counted across 1,213 of the 2,113 authors here whose files we hold, read 2026-09-06
- Cite sources for every important claimin 47 of 1213, across 38 files
- Separate facts from inferences and recommendationsin 21 of 1213, across 12 files
- Write findings to a markdown filein 19 of 1213
- Label every insight with a confidence levelin 18 of 1213, across 8 files
- Read product marketing context before asking questionsin 18 of 1213, across 8 files
- Rank themes by frequency and intensityin 16 of 1213, across 6 files
- Establish research mode before proceedingin 16 of 1213, across 6 files
- Segment survey responses by customer tier or tenurein 16 of 1213, across 6 files
- Categorize support tickets before analyzingin 16 of 1213, across 6 files
- Weight research sources from the last twelve monthsin 16 of 1213, across 6 files
- Use at least five data points per segmentin 15 of 1213, across 5 files
- Extract verbatim quotes for all research findingsin 15 of 1213, across 5 files
Said here and by no other author read
- narrow failure to smallest reproducible case
- list and rank likely causes
- design tests to confirm or eliminate hypotheses
- add strategic debug logging
- add a test to prevent recurrence
- confirm fix with manual verification
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.