Inspect Scout

Evals

Tool (not an eval — tooling/reference entry)

Transcript analysis: scan large volumes of agent eval logs for behaviors, failures, cheating.

Category
AI control & monitoring
Insurance exposure
Model failure
Publisher
Meridian Labs
Publisher type
Nonprofit / research org
License
MIT
Status
Active (<90d)
Last commit
2026-10-15
Stars
82
Notes
Meridian Labs is the nonprofit that took over Petri from Anthropic (May 2026).

View source on GitHub →


Wanich Risk Lab, 2026.