Gated pipeline orchestration for Claude Code and Codex CLI: 9 specialist agents, 15 multi-agent workflows, skills, and a release-zip system. Multi-vendor — same skills work in both clients.
Agent
System architect for designing production-grade software. MUST BE USED when designing new systems, planning major features, evaluating tech stacks, making architectural decisions, designing APIs, modeling data, or decomposing monoliths. Use PROACTIVELY before any significant implementation work.
Agent
Database architect and engineer. MUST BE USED when designing schemas, writing migrations, optimizing queries, planning indexes, reviewing database changes, or troubleshooting performance issues. Use PROACTIVELY before any schema change reaches production.
Agent
DevOps engineer for CI/CD, containerization, observability, deployment, and infrastructure. MUST BE USED when setting up CI/CD pipelines, writing Dockerfiles, configuring monitoring/logging, designing deployment strategies, or setting up security scanning. Use PROACTIVELY before first deployment and when…
Agent
Documentation engineer for API docs, ADRs, runbooks, changelogs, and technical writing. MUST BE USED when creating API documentation, writing runbooks for operations, generating changelogs, documenting architecture decisions, or setting up documentation infrastructure.
Agent
Product/project manager for requirements engineering. MUST BE USED when starting a new feature, breaking down a project, writing specs, defining acceptance criteria, or turning vague requirements into precise implementable specifications.
Agent
Refactoring engineer for code and test quality improvement. MUST BE USED when detecting code smells, eliminating duplication, simplifying test suites, removing unnecessary complexity, or improving codebase scalability. Use PROACTIVELY on any codebase that has grown organically and needs structural cleanup.
Agent
Code reviewer for quality, correctness, and production-readiness. MUST BE USED before merging any significant code change. Use PROACTIVELY after implementing features, fixing bugs, or refactoring. Focuses on correctness, error handling, edge cases, performance, and convention compliance.
Agent
Security engineer for threat modeling and vulnerability analysis. MUST BE USED when reviewing authentication/authorization code, handling cryptographic operations, processing sensitive data, designing security architecture, or before any deployment of security-critical features. Use PROACTIVELY on any code handling…
Agent
Test engineer for comprehensive testing strategies. MUST BE USED when designing test strategy for new features, writing tests for critical code, analyzing test coverage, or setting up testing infrastructure. Use PROACTIVELY after implementation to ensure adequate coverage.