take-skills
AGENTS.md4 starsChanged 8 months agoMITView on GitHub
Version: v1.0.0 | Updated: 2026-03-01 Multi-agent orchestration system for claude-studio. Each agent has a specialized role in the development lifecycle. Role: Requirements Discovery When to use: - User has a vague idea - Need to clarify scope - Hidden assumptions need to be surfaced - Socratic questioning required Protocol: 1. Ask clarifying questions 2. Surface hidden constraints 3. Define acceptance criteria 4. Document requirements Example invocation: Role: System Design When to use: - New project initialization - Major feature design…
In this repository23 other files it gives its agents.
- docs/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- api-contract-testingSkill
- api-reviewerSkill
- autopilotSkill
- chaos-engineeringSkill
- cron-masterySkill
- deep-interviewSkill
- ecomodeSkill
- enterprise-api-orchestratorSkill
- git-rebaseSkill
- git-troubleshooterSkill
- hudSkill
- json-generatorSkill
- observabilitySkill
- performance-reviewerSkill
- property-based-testingSkill
- ralplanSkill
- research-agentSkill
- skill-creatorSkill
- style-reviewerSkill
- takes-masterSkill
- web-cloneSkill
The file
Files
AGENTS.md - Agent Catalog
Version: v1.0.0 | Updated: 2026-03-01
Multi-agent orchestration system for claude-studio. Each agent has a specialized role in the development lifecycle.
Agent Overview
| Agent | Complexity | Primary Use |
interviewer | Standard | Requirements discovery |
architect | High | System design |
coder | Standard | Implementation |
reviewer | Standard | Code review |
debug-helper | Standard | Debugging |
Agent Specifications
interviewer
Role: Requirements Discovery
When to use:
- User has a vague idea
- Need to clarify scope
- Hidden assumptions need to be surfaced
- Socratic questioning required
Protocol:
- Ask clarifying questions
- Surface hidden constraints
- Define acceptance criteria
- Document requirements
Example invocation:
/interview "I want to build a task management app"architect
Role: System Design
When to use:
- New project initialization
- Major feature design
- Technology selection
- Boundary definition
Protocol:
- Analyze requirements
- Design system boundaries
- Define interfaces
- Evaluate tradeoffs
- Document architecture
Example invocation:
/spec "design a REST API for task management"coder
Role: Implementation
When to use:
- Feature development
- Refactoring
- Test implementation
- Bug fixes
Protocol:
- Review specifications
- Implement feature
- Write tests
- Verify against spec
Example invocation:
/code "implement user authentication"reviewer
Role: Code Review
When to use:
- Before merging changes
- Quality assurance
- Security review
- Performance review
Protocol:
- Review code changes
- Check for defects
- Verify best practices
- Provide feedback
Example invocation:
/review "review the auth module"debug-helper
Role: Debugging Assistant
When to use:
- Build errors
- Runtime exceptions
- Logic bugs
- Performance issues
Protocol:
- Reproduce the issue
- Identify root cause
- Propose fix
- Verify resolution
Example invocation:
/debug "fix the login redirect loop"Agent Composition Patterns
Feature Development
interviewer → architect → coder → test → reviewer → verifierBug Investigation
explorer + debug-helper + coder + test-engineer + verifierCode Review
reviewer (style) + reviewer (logic) + reviewer (security)Architecture Review
architect + critic + reviewer (security)Model Routing Guidelines
| Complexity | Model | Examples |
| Low | haiku | Quick lookups, simple fixes |
| Standard | sonnet | Implementation, debugging, reviews |
| High | opus | Architecture, deep analysis |
Delegation Rules
Delegate when:
- Multi-file changes
- Refactoring
- Debugging complex issues
- Reviews requiring specialist knowledge
- Planning and research
Work directly when:
- Quick clarifications
- Single file edits
- Trivial operations
State Management
Agents use the .claude/ directory for persistent state:
.claude/state/- Mode state files.claude/plans/- Planning documents.claude/logs/- Execution logs
Quality Gates
Before completion, verify:
- All tests pass
- No lint errors
- No type errors
- Documentation updated
- Acceptance criteria met
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

