A persistent epistemic substrate for AI. Nous treats language models as larynx, not mind, and benchmarks epistemic structure instead of output fluency.
Local-first epistemic knowledge substrate for AI agents — provenance, contention, stigmergic refinement, MCP-native
