Visual low-code tool for building and deploying AI agents and workflows.
Warden evaluates AI-agent projects across 17 governance dimensions: tool-call enforcement, agent identity, human-in-the-loop, audit trails, credential management, supply-chain security, trap defense, and more. Scores are normalized to 0–100 from 235 raw points. See the full methodology for how each dimension is calculated.
A score below 33 is UNGOVERNED, 33–59 is AT_RISK, 60–79 is PARTIAL, and 80+ is GOVERNED. Most general-purpose AI frameworks land in the AT_RISK or PARTIAL tiers — that's expected, because they are libraries, not governance platforms. The score reflects the governance posture of the framework's own source, not of any given application built on top of it.
This is an automated scan of open-source code. Warden is vendor-neutral and maintains a public scoring model. Framework maintainers are welcome to open an issue with corrections.