An AI-powered Claude Code plugin that transforms a product idea into a complete, research-backed PRD with modern architecture design, enforcement skills, and self-evolving project documentation — ready for AI-assisted development.
Use this agent to design a complete technical architecture for a product, including stack selection, system diagrams, ER model, and Architecture Decision Records. Chooses the most modern, compatible, and product-fit stack — not limited by user's current knowledge. Also handles incremental architecture updates in…
Use this agent to generate or update CI/CD pipeline files based on VCS choice (GitHub/GitLab/Bitbucket) and current architecture stack. Generates lean, essential pipelines with path-filtering per service for monorepos. Operates in full (initial) or update (evolution) mode.
Use this agent when you need to research technical specifications, regulatory requirements, API capabilities, or any factual information about a technology or domain. ONLY searches official sources. Triggers automatically when a technology or regulated domain is identified during product interrogation.
Use this agent to write or update the PRD.md, ARCHITECTURE.md, and ER.md files from structured requirements and architecture data. All output is in English. Handles both initial generation and incremental section updates for /prd-evolve.
Use this agent when you need to analyze a partial product context packet and identify what critical information is still missing, generate domain-specific follow-up questions, or detect technologies and regulated domains that require official research. Triggers after each interrogation block in /prd-new.
Use this agent to transform a completed product context packet and validated research into structured functional requirements, non-functional requirements, domain rules, compliance requirements, and acceptance criteria. Also handles delta analysis for /prd-evolve to determine which artifacts are affected by a scope…
Use this agent to independently validate research output from official-researcher agents. Must always be invoked in a fresh Agent context to prevent confirmation bias. Verifies source officiality, information accuracy, recency, and internal consistency. Triggers after each research batch in /prd-new and /prd-evolve.
Use this agent to generate or update the seven project-specific enforcement skills (project-guardian, project-architecture, project-domain-rules, project-compliance, project-docs-stack, project-cicd, project-docs) based on the finalized PRD and architecture. All skills are written in English and include a versioned…
Use this agent to generate stack-specific CLAUDE.md development guides based on the chosen architecture. Each guide applies Clean Code, Clean Architecture, DDD, The Pragmatic Programmer, and stack-specific official best practices. All output in English.