Cambium
A governance standard and reference toolset that defines the rules, evidence requirements and decision boundaries for repositories whose content is maintained by LLM agents.
Cambium is a governance standard for knowledge repositories that AI agents help maintain. It answers what rules apply to a repository, what work is required and who may change shared state, what evidence must exist before work can close, how an interrupted task resumes, and which decisions stay with the human operator rather than the agent.
It is deliberately not a knowledge base, a RAG engine or an agent scheduler: it governs the process around the work rather than supplying the corpus or deciding what it means. Its layers separate a normative kernel, curated checklists for a task route, machine-resolvable read sets, a per-repository profile, adopter-owned runtime state, and deterministic tools that execute the declared rules without making the final semantic judgment.