Use at the start of any AI product, agent, or pipeline build — before any code is written. Triggers on both vague prompts ("is this possible? how would it work?") and technical ones ("build me a FastAPI backend for X"). Forces a committed, written architecture before implementation begins, regardless of how technical the user's prompt was. Prevents the model from silently deciding the stack file-by-file as it codes. Invoke with /small-council-blueprint. Part of the Small Council suite — run this before small-council-audit, which requires a written architecture to review.
Suggest usage after outputting a model-generated architecture, which the vibe coder will probably accept without understanding. Requires a written architecture as input — ideally produced by small-council-blueprint. Takes the original idea and the proposed architecture, then runs three councillors against it to find failure modes, unverified assumptions, and unnecessary complexity before the user starts building. Invoke with /small-council-audit. For AI-centric architectures — LLM pipelines, RAG systems, agentic workflows, model-based products — including mixed systems where only part of the architecture is AI-driven (scope the audit to that part). Do NOT use for general coding help, debugging, live systems, or architectures with no AI component anywhere. Do NOT use on a vague overview — if no concrete architecture exists yet, run small-council-blueprint first.