v1.2.62026-06-17

Adjudication-engine agent contract, live doctor, contract eval

An agent-experience release. The agent contract now teaches Myco as the adjudication engine it is, and the doctor verifies your setup for real instead of trusting config. No tool-contract changes (brain_ingest gains an additive extraction field); no migrations.

Changed

The agent contract embodies the thesis. The program writes the facts, not the model. The contract now teaches the source-first write ladder (ingest a source, then propose a claim, then a private save_memory) and the engine mechanics: compounding confidence across independent sources, supersede-and-never-overwrite on conflict, provenance via brain_why, and propose to review to promote.
save_memory honestly fenced. brain_save_memory is a private scratchpad, never workspace truth. The contract is single-sourced across the runtime, the pasteable CLAUDE.md / .cursorrules / AGENTS.md manual, the tool descriptions, and the website, with a CI drift guard so the surfaces cannot diverge.

Added

mycobrain-doctor verifies live. For the local Ollama path it pings the server, confirms the model is pulled, and runs a real embed and generation, then reports live-verified. It adds an extraction-backlog check and a --fix mode that pulls missing models. brain_ingest returns an honest extraction receipt (graph vs search-only) so an agent knows whether a fact graph was actually built.
Agent-contract eval harness. npm run eval:contract scores whether an agent routes to the right tool across 20 adversarial scenarios, so the contract is measured and regression-guarded.

No breaking changes to any brain_* tool contract. Full details in the repository CHANGELOG.