CtrlK
BlogDocsLog inGet started
Tessl Logo

investigate

Systematic debugging with Iron Law methodology. 5-phase investigation from evidence collection to verified fix. Triggers on "investigate", "debug", "root cause".

57

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/investigate/SKILL.md
SKILL.md
Quality
Evals
Security

Investigate

Systematic debugging skill that enforces the Iron Law: never fix without a confirmed root cause. Produces a structured debug report with full evidence chain, minimal fix, and regression test.

Iron Law Principle

No fix without confirmed root cause. Every investigation follows a strict evidence chain:

  1. Reproduce the bug with concrete evidence
  2. Analyze patterns to assess scope
  3. Form and test hypotheses (max 3 strikes)
  4. Implement minimal fix ONLY after root cause is confirmed
  5. Verify fix and generate structured report

Violation of the Iron Law (skipping to Phase 4 without Phase 3 confirmation) is prohibited.

Key Design Principles

  1. Evidence-First: Collect before theorizing. Logs, stack traces, and reproduction steps are mandatory inputs.
  2. Minimal Fix: Change only what is necessary. Refactoring is not debugging.
  3. 3-Strike Escalation: If 3 consecutive hypothesis tests fail, STOP and escalate with a diagnostic dump.
  4. Regression Coverage: Every fix must include a test that fails without the fix and passes with it.
  5. Structured Output: All findings are recorded in machine-readable JSON for future reference.

Execution Flow

Phase 1: Root Cause Investigation
  Reproduce bug, collect evidence (errors, logs, traces)
  Use ccw cli --tool gemini --mode analysis for initial diagnosis
  Output: investigation-report.json
      |
      v
Phase 2: Pattern Analysis
  Search codebase for similar patterns (same error, module, antipattern)
  Assess scope: isolated vs systemic
  Output: pattern-analysis section in report
      |
      v
Phase 3: Hypothesis Testing
  Form max 3 hypotheses from evidence
  Test each with minimal read-only probes
  3-strike rule: STOP and escalate on 3 consecutive failures
  Output: confirmed root cause with evidence chain
      |
      v
Phase 4: Implementation  [GATE: requires Phase 3 confirmed root cause]
  Implement minimal fix
  Add regression test
  Verify fix resolves reproduction case
      |
      v
Phase 5: Verification & Report
  Run full test suite
  Check for regressions
  Generate structured debug report to .workflow/.debug/

Directory Setup

mkdir -p .workflow/.debug

Output Structure

.workflow/.debug/
  debug-report-{YYYY-MM-DD}-{slug}.json    # Structured debug report

Completion Status Protocol

This skill follows the Completion Status Protocol defined in _shared/SKILL-DESIGN-SPEC.md sections 13-14.

StatusWhen
DONERoot cause confirmed, fix applied, regression test passes, no regressions
DONE_WITH_CONCERNSFix applied but partial test coverage or minor warnings
BLOCKEDCannot reproduce bug, or 3-strike escalation triggered in Phase 3
NEEDS_CONTEXTMissing reproduction steps, unclear error conditions

Reference Documents

DocumentPurpose
phases/01-root-cause-investigation.mdEvidence collection and reproduction
phases/02-pattern-analysis.mdCodebase pattern search and scope assessment
phases/03-hypothesis-testing.mdHypothesis formation, testing, and 3-strike rule
phases/04-implementation.mdMinimal fix with Iron Law gate
phases/05-verification-report.mdTest suite, regression check, report generation
specs/iron-law.mdIron Law rules definition
specs/debug-report-format.mdStructured debug report JSON schema

CLI Integration

This skill leverages ccw cli for multi-model analysis at key points:

PhaseCLI UsageMode
Phase 1Initial diagnosis from error evidence--mode analysis
Phase 2Cross-file pattern search--mode analysis
Phase 3Hypothesis validation assistance--mode analysis

All CLI calls use --mode analysis (read-only). No write-mode CLI calls during investigation phases 1-3.

Repository
catlog22/Claude-Code-Workflow
Last updated
First committed

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.