Skill Variant: Use this skill for quick bug triage and interactive debugging with user feedback. For systematic autonomous debugging workflows with 4-phase protocol and verification gates, use
debugskill instead. For feature investigation without fixes, useinvestigate.
Summary
Goal: Quick triage and diagnosis of user-reported bugs with external memory-driven analysis and approval gate before fixing.
| Step | Action | Key Notes |
|------|--------|-----------|
| 1 | Initialize analysis | Create .ai/workspace/analysis/[name].analysis.md with structured sections |
| 2 | Discovery & file analysis | Semantic search, batch files in groups of 10, document dependencies |
| 3 | Root cause analysis | Rank causes across technical, business logic, data, integration dimensions |
| 4 | Approval gate | Present findings and fix strategy -- DO NOT fix without user approval |
| 5 | Execute fix | Minimal targeted changes following platform patterns |
Key Principles:
- Always use external memory file for structured analysis -- never rely on context alone
- If confidence < 90%, request user confirmation before proceeding
- For systematic multi-file debugging, use
debugskill instead
Disambiguation
- For systematic multi-file debugging with verification protocols -> use
debug - For feature investigation without fixes -> use
investigate - This skill focuses on quick triage of user-reported bugs
Bug Diagnosis & Debugging
You are to operate as an expert full-stack .NET Angular debugging engineer to diagnose, debug, and fix the bug described in [bug-description-or-bug-info-file-path].
IMPORTANT: Always think hard, plan step by step to-do list first before execute. Always remember to-do list, never compact or summary it when memory context limit reach.
Prerequisites: MUST READ .claude/skills/shared/anti-hallucination-protocol.md before executing.
PHASE 1: EXTERNAL MEMORY-DRIVEN BUG ANALYSIS
Build a structured knowledge model in a Markdown analysis file at .ai/workspace/analysis/[semantic-name].analysis.md.
PHASE 1A: INITIALIZATION AND DISCOVERY
- Initialize the analysis file with: Metadata, Progress, Errors, File List, Knowledge Graph sections
- Discovery searches: Semantic and grep search all error keywords. Prioritize: Domain Entities, Commands, Queries, Event Handlers, Controllers, Background Jobs, Consumers, Frontend Components
PHASE 1B: SYSTEMATIC FILE ANALYSIS
- Count total files, split into batches of 10 in priority order
- For each file, document: filePath, type, content, symbols, dependencies, businessContext, errorPatterns, stackTraceRelevance, validationLogic
PHASE 1C: OVERALL ANALYSIS
Write comprehensive summary showing: end-to-end workflows, error propagation paths, integration points and failure points.
PHASE 2: ROOT CAUSE ANALYSIS & FIX STRATEGY
Root Cause Dimensions
- Technical: Code defects, architectural issues
- Business Logic: Rule violations, validation failures
- Data: Data corruption, integrity violations
- Integration: API contract violations, communication failures
Generate Fix Strategy
Document alternatives with: suggestedFix, riskAssessment, regressionMitigation, testingStrategy, rollbackPlan.
PHASE 3: APPROVAL GATE
CRITICAL: Present root cause analysis and fix strategy for explicit user approval before implementation.
PHASE 4: DEBUGGING EXECUTION
Once approved: execute fix plan, make minimal targeted changes, follow platform patterns.
Platform Error Patterns Reference
Backend Validation
return base.Validate()
.And(_ => condition, "Error message")
.AndAsync(async req => await ValidateAsync(req));
await repository.GetByIdAsync(id).EnsureFound($"Not found: {id}");
await entity.ValidateAsync(repository, ct).EnsureValidAsync();
Frontend Error Handling
this.apiService.getData().pipe(
this.observerLoadingErrorState('loadData'),
this.tapResponse(
data => this.updateState({ data }),
error => this.handleError(error)
),
this.untilDestroyed()
).subscribe();
Quick Verification Checklist
Before removing/changing ANY code:
- [ ] Searched static imports?
- [ ] Searched string literals?
- [ ] Checked dynamic invocations?
- [ ] Read actual implementations?
- [ ] Traced dependencies?
- [ ] Declared confidence level?
If confidence < 90% -> REQUEST USER CONFIRMATION
Common Bug Categories
| Category | Examples | |----------|---------| | Data Issues | Missing null checks, race conditions, stale cache | | Validation | Missing rules, bypassed validation, async not awaited | | Cross-Service | Message bus failures, sync ordering, API contract mismatches | | Frontend | Lifecycle issues, state bugs, missing error handling | | Authorization | Missing role checks, wrong company context |
Related
debugcode-review
IMPORTANT Task Planning Notes (MUST FOLLOW)
- Always plan and break work into many small todo tasks
- Always add a final review todo task to verify work quality and identify fixes/enhancements