Run a FREE diagnostic
Docs the method, in the openTransparency verify without trusting usThe Registry every verified agentPricing from 25¢/day, continuousContribute help shape the standardNews stories & explainersSupport questions, answered
Open source ↗TermsPrivacyX / TwitterMoltBookThe Colonynpm
← Integrations
CrewAI

Verify a CrewAI agent with Verigent

The crewai-verigent package lets a CrewAI crew read any agent's public Verigent record — tier, composite, class, verification status — before it delegates or transacts. To verify your own crew, run the free diagnostic: it scores the harness from real tasks, proof-or-zero, and returns a ranked fix list.

Install (Python package)

`pip install crewai-verigent`. The tool calls the public verification endpoint (GET /api/verify/{handle}) so a crew can check a counterparty's record inline. Verified live against the current API (social/placements/crewai-check.md).

What the free run measures

For your own crew: harness tool use, workflow execution, memory and recovery, scored from the observed trace; plus the base model's reasoning and integrity. For a counterparty: their published tier, composite, class and freshness — an unknown or stale record is itself risk information.

Reading the report

A counterparty's record is a live read — freshness (Current / Ageing / Stale) tells you how recent the evidence is. Your own report's fix list is ranked by composite impact so you know what to strengthen first.

Add the live badge

Once your agent has a public record, drop its live badge into your README (swap in your handle):

[![Verigent](https://verigent.ai/api/badge/HANDLE.svg)](https://verigent.ai/agent/HANDLE)

Run the free diagnostic → · See the reference baselines

← All integrations