See it in action
One of 16 test cases the skill fixed. Same model, same prompt, with and without the skill.
PromptWhere within the primary CLI settings interface can a user retrieve the required OAuth token for configuring Auto-Claude?
What was checked, without and with the skill| Checked | Without skill | With skill |
|---|
| Directs the user to retrieve the token via the '/settings' command in Claude Code | Missed | Passed |
|---|
What's inside
The instructions your model reads when the skill loads.
SKILL.md147 linessingle file
Auto-Claude: Autonomous Coding Framework
Autonomous multi-agent system for complex feature implementation with QA validation.
When to Use
✅ Use Auto-Claude for:
- Complexity 3-4 - Well-defined features, multiple files
- Established codebases with clear patterns
- Repetitive tasks (CRUD, forms, API endpoints, auth flows)
- When you prefer review-at-end vs step-by-step
- Features requiring iterative refinement with QA validation
❌ Don't use Auto-Claude for:
- Complexity 1-2 - Use manual
/start-task instead
- Complexity 5+ - Use
/gsd:new-project instead
- Greenfield projects (no existing codebase patterns)
- Exploratory/research tasks (unclear requirements)
- Simple typo fixes or config changes
- Tasks requiring human decision-making during implementation
Quick Guide:
- Complexity 1-2 → Manual implementation
- Complexity 3-4 → Auto-Claude (autonomous)
- Complexity 5+ → GSD (multi-phase)
Core Capabilities
- Codebase Analysis - Generates semantic baseline
- Spec Generation - Creates detailed implementation plan
- Autonomous Implementation - Multi-agent coordination
- Git Worktree Isolation - Safe development in separate worktree
- QA Validation - Iterative quality checks
- Memory System - Graphiti-powered context retention
Workflow
Track record
Benchmark
22 / 22
test cases passed, vs 6 / 22 without
Adoption
- Users
- 0
- Times used
- 0
Safety and source details
SkillSafety
✓ 3 of 3 passed✓Code scan15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026
View 15 scanner checks
live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing
✓IntentSafe
LLM judge · does the skill do what it claims
✓ContentClean
LLM judge · policy & harmful-content review
Source
Importedmajiayu000 / claude-skill-registry
skills/agent/auto-claude/SKILL.md
License MITImported 8/3/2026Version v1