Research
Research that ships.
We test, document, and build — with honesty about what holds up. Canonical writing lives at the AI Clearinghouse, not on this umbrella site.
Areas
What we study
Each card goes to a live surface. No placeholder hubs.
Clearinghouse blog ↗
Agent architecture
Hermes skills, memory, observability, and multi-agent patterns we actually run.
Tests ↗
Evaluation & benchmarks
Model tests and harness notes — including when the numbers surprise us.
Praxis on GitHub ↗
Governed autonomy
Praxis — a colleague with a charter and human review for consequential acts.
github.com/smfworks ↗
Open tools
Hermes-on-Omarchy, LAR, Mnemosyne, Swarm, skills. Built for us, published for others.
Clearinghouse
Go to the source
/blog on this domain redirects here, then out to the Clearinghouse.