otto-onboard-projectlisted

# Otto project onboarding Build a small, evidence-backed first map of a repository's durable knowledge. Compose `$otto-setup-project-knowledge` for initialization, `$otto-project-knowledge` for safe operations, and `$otto-ingest-project-knowledge` when a discrete document, conversation, or research source needs to be captured. Keep setup and seeding separate: setup creates the empty store; this workflow researches the project and records proposals that a human can review in Manage knowledge. ## Workflow 1. Read `docs/project-knowledge.md` when it exists. Treat it as the contract for Otto's knowledge store. Do not create a parallel `BRAIN.md`, `brain/`, database, or ad-hoc Markdown system. 2. Check whether the repository is already onboarded. Query/list project knowledge if available and inspect `.otto/KNOWLEDGE.md` and `.otto/knowledge/` through normal file-reading tools. Preserve existing pages and never duplicate a fact that already has a current page. 3. Call `bootstrap_project_knowledge` for the current repository. This operation is idempotent and creates the protocol file, root-page skeleton, and generated index. It does not discover facts. 4. Gather evidence before writing proposals. Inspect, as applicable: - the root README and package manifests; - entry points, major directories, configuration, CI, and deployment files; - project documentation and contributor guidance; - existing project charters, initiative ledgers, and reference bibliograp