# CVE-2026-92206: CrewAI Framework unsafe reflection in agent loading remote code execution

Source: https://junglewise.ai/threats/cve-2026-92206-crewai-crewai-framework-agent-loading-unsafe-reflection-remote  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-10-21. Updated: 2026-10-21  
License: CC BY 4.0 with a link back

CVE-2026-92206 · Severity: high · CVSS 8.8 · Published 2026-10-21

CrewAI is a framework for building autonomous agent systems. A vulnerability in its agent loading mechanism allows remote attackers to execute arbitrary code on systems running the framework if a user loads a malicious agent configuration. An attacker can gain code execution with the privileges of the service account running CrewAI.

Vendors: [crewAI](https://junglewise.ai/threats/vendors/crewai).

## Quick answers

**Is CVE-2026-92206 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-92206 being exploited in the wild.

**How severe is CVE-2026-92206?** High severity, with a CVSS score of 8.8, published 21 October 2026.

**Which products and versions are affected?** CrewAI CrewAI.

**Is there a fix for CVE-2026-92206?** No fixed version is recorded yet. Restrict user interaction with the product by preventing the loading of agent configurations from untrusted sources. If you run it, Junglewise drafts a mitigation that fits your setup and applies it once you approve, with senior security engineers behind it.

**How is CVE-2026-92206 attacked?** It is attacked over the network, in a software library.

**Am I affected by CVE-2026-92206?** Check whether you run CrewAI. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## How to fix


Restrict user interaction with the product by preventing the loading of agent configurations from untrusted sources.

## Technical details

The vulnerability exists in the load_agent_from_repository function, which fails to properly validate user-supplied arguments before using them to dynamically import modules via unsafe reflection. An attacker can craft a malicious agent configuration that, when loaded by a user, triggers arbitrary code execution in the context of the service account. Exploitation requires user interaction—the target must explicitly load the attacker's configuration.

## Affected products

- CrewAI CrewAI

## Timeline

- 2025-10-29: disclosed: Vulnerability reported to vendor
- 2026-09-16: advisory: Coordinated public release of advisory

## References

- [https://www.zerodayinitiative.com/advisories/ZDI-26-706/](https://www.zerodayinitiative.com/advisories/ZDI-26-706/)

## Related threats

- [CrewAI server-side request forgery in ScrapWebsiteTool](https://junglewise.ai/threats/cve-2026-120630-crewai-contains-an-ssrf-vulnerability-caused-by-its-ssrf-guard)