Key Features
Tech Stack
Most agent frameworks treat the LLM like a “smart intern” — I wanted to treat it like a superintelligence that needs to be kept in a constitutional prison with multiple immune systems
The entire pipeline is deliberately rigid and deterministic on purpose: every single decision has to go through Evidence → Critique → Debate → ValueCore → FUJI Gate → hash-chained TrustLog. No shortcuts, no “hope the model behaves”.
Would love to hear your thoughts once you dig in — especially on whether the safety overhead feels worth it in practice.
(Also happy to answer any questions about the Doctor Dashboard or the Telos two-axis value system!)
Not affiliated with Hacker News or Y Combinator. We simply enrich the public API with analytics.