Provides tools for agents to manage a local review graph, tracking acceptance behaviors, evidence, review passes, and human waivers to decouple review convergence from shipping readiness.
Provides MCP tools for AI agents to classify changes into autonomous or human review lanes, and to run shadow simulations over backlogs without shipping anything.
Safety checks an agent runs before it acts — flags malicious packages, destructive shell commands, secret leaks, and web-backend holes before execution.
Evaluates on-chain risk for Pharos agents before executing transactions, providing verdicts (safe/caution/dangerous) and risk-bounded execution plans via Foundry cast reads.