By Vero SecurityBuilt and funded by Vero Security, the AI Security Engineering platform.Read the charter

Baselines and drift checks

A written baseline per agent, covering mode, rules, sandbox, egress, MCP and hooks, gives auditors and engineers the same target. Rerun inventory queries to catch drift such as a bypass mode set in a user file or a newly added MCP server. Published baselines and threat taxonomies are a good place to start.

Read in this order

  1. Open sourcetrailofbits/claude-code-configStart here: a real opinionated baseline you can fork and adjust.
  2. OfficialOWASP Top 10 for Agentic Applications for 2026A threat list to check your baseline against for gaps.
  3. Video200 Bugs/Week/Engineer: How We Rebuilt Trail of Bits Around AI ([un]prompted 2026)Shows shared skills, sandbox patterns and configuration across a firm. 30 min.

Save your progress

We email you a sign-in link. No password.

By signing in you agree to the terms. The privacy policy explains how we use your details.

Join the AI security engineering community

Free. Roadmap updates, chapters near you, and first word when credentials open.

We’ll email you community updates. Unsubscribe anytime. By sending this you agree to the terms. The privacy policy explains how we use your details.