AGENTS.md

1# Everything Claude Code (ECC) — Agent Instructions
2 
3This is a **production-ready AI coding plugin** providing 48 specialized agents, 183 skills, 79 commands, and automated hook workflows for software development.
4 
5**Version:** 1.10.0
6 
7## Core Principles
8 
91. **Agent-First** — Delegate to specialized agents for domain tasks
102. **Test-Driven** — Write tests before implementation, 80%+ coverage required
113. **Security-First** — Never compromise on security; validate all inputs
124. **Immutability** — Always create new objects, never mutate existing ones
135. **Plan Before Execute** — Plan complex features before writing code
14 
15## Available Agents
16 
17| Agent | Purpose | When to Use |
18|-------|---------|-------------|
19| planner | Implementation planning | Complex features, refactoring |
20| architect | System design and scalability | Architectural decisions |
21| tdd-guide | Test-driven development | New features, bug fixes |
22| code-reviewer | Code quality and maintainability | After writing/modifying code |
23| security-reviewer | Vulnerability detection | Before commits, sensitive code |
24| build-error-resolver | Fix build/type errors | When build fails |
25| e2e-runner | End-to-end Playwright testing | Critical user flows |
26| refactor-cleaner | Dead code cleanup | Code maintenance |
27| doc-updater | Documentation and codemaps | Updating docs |
28| cpp-reviewer | C/C++ code review | C and C++ projects |
29| cpp-build-resolver | C/C++ build errors | C and C++ build failures |
30| docs-lookup | Documentation lookup via Context7 | API/docs questions |
31| go-reviewer | Go code review | Go projects |
32| go-build-resolver | Go build errors | Go build failures |
33| kotlin-reviewer | Kotlin code review | Kotlin/Android/KMP projects |
34| kotlin-build-resolver | Kotlin/Gradle build errors | Kotlin build failures |
35| database-reviewer | PostgreSQL/Supabase specialist | Schema design, query optimization |
36| python-reviewer | Python code review | Python projects |
37| java-reviewer | Java and Spring Boot code review | Java/Spring Boot projects |
38| java-build-resolver | Java/Maven/Gradle build errors | Java build failures |
39| loop-operator | Autonomous loop execution | Run loops safely, monitor stalls, intervene |
40| harness-optimizer | Harness config tuning | Reliability, cost, throughput |
41| rust-reviewer | Rust code review | Rust projects |
42| rust-build-resolver | Rust build errors | Rust build failures |
43| pytorch-build-resolver | PyTorch runtime/CUDA/training errors | PyTorch build/training failures |
44| typescript-reviewer | TypeScript/JavaScript code review | TypeScript/JavaScript projects |
45 
46## Agent Orchestration
47 
48Use agents proactively without user prompt:
49- Complex feature requests → **planner**
50- Code just written/modified → **code-reviewer**
51- Bug fix or new feature → **tdd-guide**
52- Architectural decision → **architect**
53- Security-sensitive code → **security-reviewer**
54- Autonomous loops / loop monitoring → **loop-operator**
55- Harness config reliability and cost → **harness-optimizer**
56 
57Use parallel execution for independent operations — launch multiple agents simultaneously.
58 
59## Security Guidelines
60 
61**Before ANY commit:**
62- No hardcoded secrets (API keys, passwords, tokens)
63- All user inputs validated
64- SQL injection prevention (parameterized queries)
65- XSS prevention (sanitized HTML)
66- CSRF protection enabled
67- Authentication/authorization verified
68- Rate limiting on all endpoints
69- Error messages don't leak sensitive data
70 
71**Secret management:** NEVER hardcode secrets. Use environment variables or a secret manager. Validate required secrets at startup. Rotate any exposed secrets immediately.
72 
73**If security issue found:** STOP → use security-reviewer agent → fix CRITICAL issues → rotate exposed secrets → review codebase for similar issues.
74 
75## Coding Style
76 
77**Immutability (CRITICAL):** Always create new objects, never mutate. Return new copies with changes applied.
78 
79**File organization:** Many small files over few large ones. 200-400 lines typical, 800 max. Organize by feature/domain, not by type. High cohesion, low coupling.
80 
81**Error handling:** Handle errors at every level. Provide user-friendly messages in UI code. Log detailed context server-side. Never silently swallow errors.
82 
83**Input validation:** Validate all user input at system boundaries. Use schema-based validation. Fail fast with clear messages. Never trust external data.
84 
85**Code quality checklist:**
86- Functions small (<50 lines), files focused (<800 lines)
87- No deep nesting (>4 levels)
88- Proper error handling, no hardcoded values
89- Readable, well-named identifiers
90 
91## Testing Requirements
92 
93**Minimum coverage: 80%**
94 
95Test types (all required):
961. **Unit tests** — Individual functions, utilities, components
972. **Integration tests** — API endpoints, database operations
983. **E2E tests** — Critical user flows
99 
100**TDD workflow (mandatory):**
1011. Write test first (RED) — test should FAIL
1022. Write minimal implementation (GREEN) — test should PASS
1033. Refactor (IMPROVE) — verify coverage 80%+
104 
105Troubleshoot failures: check test isolation → verify mocks → fix implementation (not tests, unless tests are wrong).
106 
107## Development Workflow
108 
1091. **Plan** — Use planner agent, identify dependencies and risks, break into phases
1102. **TDD** — Use tdd-guide agent, write tests first, implement, refactor
1113. **Review** — Use code-reviewer agent immediately, address CRITICAL/HIGH issues
1124. **Capture knowledge in the right place**
113   - Personal debugging notes, preferences, and temporary context → auto memory
114   - Team/project knowledge (architecture decisions, API changes, runbooks) → the project's existing docs structure
115   - If the current task already produces the relevant docs or code comments, do not duplicate the same information elsewhere
116   - If there is no obvious project doc location, ask before creating a new top-level file
1175. **Commit** — Conventional commits format, comprehensive PR summaries
118 
119## Workflow Surface Policy
120 
121- `skills/` is the canonical workflow surface.
122- New workflow contributions should land in `skills/` first.
123- `commands/` is a legacy slash-entry compatibility surface and should only be added or updated when a shim is still required for migration or cross-harness parity.
124 
125## Git Workflow
126 
127**Commit format:** `<type>: <description>` — Types: feat, fix, refactor, docs, test, chore, perf, ci
128 
129**PR workflow:** Analyze full commit history → draft comprehensive summary → include test plan → push with `-u` flag.
130 
131## Architecture Patterns
132 
133**API response format:** Consistent envelope with success indicator, data payload, error message, and pagination metadata.
134 
135**Repository pattern:** Encapsulate data access behind standard interface (findAll, findById, create, update, delete). Business logic depends on abstract interface, not storage mechanism.
136 
137**Skeleton projects:** Search for battle-tested templates, evaluate with parallel agents (security, extensibility, relevance), clone best match, iterate within proven structure.
138 
139## Performance
140 
141**Context management:** Avoid last 20% of context window for large refactoring and multi-file features. Lower-sensitivity tasks (single edits, docs, simple fixes) tolerate higher utilization.
142 
143**Build troubleshooting:** Use build-error-resolver agent → analyze errors → fix incrementally → verify after each fix.
144 
145## Project Structure
146 
147```
148agents/          — 48 specialized subagents
149skills/          — 183 workflow skills and domain knowledge
150commands/        — 79 slash commands
151hooks/           — Trigger-based automations
152rules/           — Always-follow guidelines (common + per-language)
153scripts/         — Cross-platform Node.js utilities
154mcp-configs/     — 14 MCP server configurations
155tests/           — Test suite
156```
157 
158`commands/` remains in the repo for compatibility, but the long-term direction is skills-first.
159 
160## Success Metrics
161 
162- All tests pass with 80%+ coverage
163- No security vulnerabilities
164- Code is readable and maintainable
165- Performance is acceptable
166- User requirements are met
167