Agno

Agent framework · Agno · 3/7 verified · Last checked 2026-10-16

Human-in-the-loop covers user confirmation, input and admin approvals that persist for review. Tracing writes spans to your own database. Runs can be cancelled cooperatively.

Overall: C (2.00 avg, 3/7 controls verified; capped from B because B needs 4 verified)

Lock this down first

Require confirmation on write tools and turn on tracing=True in AgentOS.

Scorecard

Criterion	Rating	Notes
Permission granularity	Unverified	Not confirmed in the docs we checked.
Human approval gates	2/3	User confirmation pauses tool calls. Required approvals can be persisted for admin review. Set in code, not admin-lockable.
Sandboxing	Unverified	Not confirmed in the docs we checked.
Secrets handling	Unverified	Not confirmed in the docs we checked.
Audit logs	2/3	tracing=True writes spans for model calls and tool executions to your database.
Kill switch / interrupt	2/3	cancel_run() stops a run at the next cooperative checkpoint. An operation already in progress may finish.
Network egress control	Unverified	Not confirmed in the docs we checked.

Unverified: Permission granularity, Sandboxing, Secrets handling, Network egress control. That means we could not confirm it in public docs, not that it is missing.

Sources

Vendor? Request a correction. We review corrections within 7 days.