Best use case
fix-bug is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Fix bug command
Teams using fix-bug should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/fix-bug/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How fix-bug Compares
| Feature / Agent | fix-bug | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Fix bug command
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
---
## LIBRARY-FIRST PROTOCOL (MANDATORY)
**Before writing ANY code, you MUST check:**
### Step 1: Library Catalog
- Location: `.claude/library/catalog.json`
- If match >70%: REUSE or ADAPT
### Step 2: Patterns Guide
- Location: `.claude/docs/inventories/LIBRARY-PATTERNS-GUIDE.md`
- If pattern exists: FOLLOW documented approach
### Step 3: Existing Projects
- Location: `D:\Projects\*`
- If found: EXTRACT and adapt
### Decision Matrix
| Match | Action |
|-------|--------|
| Library >90% | REUSE directly |
| Library 70-90% | ADAPT minimally |
| Pattern exists | FOLLOW pattern |
| In project | EXTRACT |
| No match | BUILD (add to library after) |
---
## STANDARD OPERATING PROCEDURE
### Purpose
- Primary action: Fix bug command
### Trigger Conditions
- Command syntax: /fix-bug [args]
- Ensure prerequisites are met before execution.
### Inputs and Options
- Inputs: No structured parameters defined; capture user intent explicitly.
### Execution Phases
1. Review the request and confirm scope.
2. Execute the command flow.
3. Summarize outcomes and next actions.
### Success Criteria and Outputs
- Document artifacts, decisions, and follow-up actions clearly.
### Error Handling and Recovery
- If execution fails, capture the failure mode, retry with verbose context, and surface actionable remediation steps.
### Chaining and Coordination
### Memory and Tagging
- Tag session outputs with who/when/why for traceability.
### LEARNED PATTERNS (Session: 2026-01-07)
#### Asset Selection Protocol
When multiple similar assets exist (e.g., headshot.jpg vs headshot.png):
1. List all candidates with visual inspection or metadata check
2. Confirm correct asset with user before implementation
3. Document reasoning for selection
#### Layout Restoration Pattern
For "restore", "add back", or "bring back" requests:
1. FIRST: Research git history to find original implementation
```bash
git log --all --oneline -- <file>
git show <commit>:<file>
```
2. Extract working implementation patterns
3. Apply to current codebase
4. AVOID: Trial-and-error positioning attempts without historical context
#### Positioning Decision Tree
- Hero sections with text + image -> Grid-based layout (lg:grid-cols-12)
- Simple overlays -> Absolute positioning
- If >3 positioning iterations needed -> STOP and research git history or ask for design reference
#### User Frustration Signals
Phrases like "this is getting sad", "stop", "reverse all changes" indicate:
- Trigger: Immediate rollback + strategy pivot required
- Response: Research historical solutions or ask for design reference
- Never continue iterating after frustration signals
### Example Invocation
- /fix-bug example
### Output Format
- Provide a concise summary, actions taken, artifacts generated, and recommended next steps.
- Always include an explicit confidence line: "Confidence: X.XX (ceiling: TYPE Y.YY)".
- Use ceilings — inference/report: 0.70, research: 0.85, observation: 0.95, definition: 0.95.
- Keep user-facing output in plain English; reserve VCL markers for the appendix only.
Confidence: 0.86 (ceiling: observation 0.95) - SOP rewritten to Prompt-Architect pattern based on legacy command content.
---
## VCL COMPLIANCE APPENDIX (Internal Reference)
[[HON:teineigo]] [[MOR:root:PA]] [[COM:PromptArchitect]] [[CLS:ge_command]] [[EVD:-DI<gozlem>]] [[ASP:nesov.]] [[SPC:path:/commands]]
[define|neutral] CONFIDENCE_CEILINGS := {inference:0.70, report:0.70, research:0.85, observation:0.95, definition:0.95} [conf:0.9] [state:confirmed]
[direct|emphatic] L2_LANGUAGE := English; user-facing outputs exclude VCL markers. [conf:0.99] [state:confirmed]
[commit|confident] <promise>FIX_BUG_VERILINGUA_VERIX_COMPLIANT</promise> [conf:0.88] [state:confirmed]Related Skills
/*============================================================================*/
/* SKILL SKILL :: VERILINGUA x VERIX EDITION */
web-scraping
Structured data extraction from web pages using claude-in-chrome MCP with sequential-thinking planning. Focus on READ operations, data transformation, and pagination handling for multi-page extraction.
visual-testing
Screenshot-based visual comparison and regression testing using claude-in-chrome MCP. Captures, compares, and validates UI states to detect layout shifts, visual bugs, and design regressions across viewports.
reflect
Extract learnings from session corrections and patterns, update skill files with persistent memory. Implements Loop 1.5 - per-session micro-learning between execution and meta-optimization.
e2e-test
End-to-end testing workflow for validating complete user journeys through web applications using claude-in-chrome MCP. Specializes in test assertions, suite organization, evidence collection, and pass/fail reporting.
build-feature
Build feature command
browser-automation
Complex browser automation workflow using claude-in-chrome MCP with mandatory sequential-thinking planning. Use when automating multi-step web interactions, form filling, navigation sequences, or web scraping.
image-gen
Modular image generation - supports local SDXL Lightning, OpenAI DALL-E, Replicate, or custom providers
reverse-engineering-quick-triage
Fast binary analysis with string reconnaissance and static disassembly\ \ (RE Levels 1-2). Use when triaging suspicious binaries, extracting IOCs quickly,\ \ or performing initial malware analysis. Completes in \u22642 hours with automated\ \ decision gates."
reverse-engineering-firmware-analysis
Firmware extraction and IoT security analysis (RE Level 5) for routers and embedded systems. Use when analyzing IoT firmware, extracting embedded filesystems (SquashFS/JFFS2/CramFS), finding hardcoded credentials, performing CVE scans, or auditing embedded system security. Handles encrypted firmware with known decryption schemes. Completes in 2-8 hours with binwalk+firmadyne+QEMU emulation.
reverse-engineering-deep-analysis
Advanced binary analysis with runtime execution and symbolic path exploration (RE Levels 3-4). Use when need runtime behavior, memory dumps, secret extraction, or input synthesis to reach specific program states. Completes in 3-7 hours with GDB+Angr.
reconnaissance
Systematic technology and market reconnaissance for extracting actionable intelligence from repositories, papers, and competitive landscapes.