What problem does it solve? Setting up a consistent SDLC framework across repositories requires manually creating documentation trees, state files, and project context configuration, which is error-prone and quickly drifts out of sync with the framework version. ## Core Features & Use Cases - Stack Detection: Scans repository manifests to infer language, framework, build tool, and test framework, then proposes matching stack profiles for confirmation. - Idempotent Bootstrap and Upgrade: Creates the docs/ and .agent/state/ trees, fills .agent/project-context.yaml, writes the managed project-facts block into CLAUDE.md, and seeds domain knowledge files without overwriting hand-written content. - Verification and State Initialization: Validates that every placeholder is filled, every profile resolves, and writes initial state files plus a context cache so subsequent commands start warm. - Use Case: After cloning a new FastAPI microservice template, run the init command once to detect the Python/FastAPI stack, generate the full docs and state tree, and populate CLAUDE.md with real project facts derived from the actual code. ## Quick Start Ask the AI to run /init-project to bootstrap the SDLC framework in the current repository and confirm the detected stack when prompted.
