#formal-verification

Deterministic authorization + tamper-evident audit for AI agents, humans, and workloads — one principal type, safety invariants machine-checked by an SMT solver (cvc5), every decision independently verifiable offline. Pure Rust.

A law-proved port of br (beads_rust), the local-first, dependency-aware issue tracker for AI coding agents, to the Bend 2 language. Golden-tested against the original on every lane; early work in progress.

Proof-carrying compliance certificates for AI agent actions: machine-checked (Coq, axiom-free) + zero-knowledge proofs that an agent action obeyed a formal policy.

Make an AI agent's 'done' mean 'here is the command output that proves it.' Deterministic, LLM-free verification for AI coding agents: sealed plans, SHA-256 hash-chained evidence, STRIPS/PDDL reachability, and a fail-closed completion gate.

Independent Comparator/Lean audit of two OpenAI mathematical results, with bilingual evidence, scope notes and an essay on verification and explanation.

Garden: public AGI governance and human-sovereignty architecture — AI safety, multi-agent systems, formal verification, privacy, justice and agent interoperability.

Erdős similarity for distance sets: Lean 4 formalization and paper, building on OpenAI's Falconer and dyadic Erdős-similarity results