CVE-2026-62675 omnigent-ai CVE debrief

Who should care

Defenders and administrators of systems using the Omnigent framework should assess exposure and apply the patch to prevent potential code execution and data exposure. They should also review compensating controls for exposed systems while remediation is scheduled and verified, and monitor for suspicious activity and potential code execution attempts.

Why it matters

CVE-2026-62675 is a high-severity vulnerability in the Omnigent AI agent framework that allows an authenticated user to execute a local command with runner process permissions. Defenders should prioritize verifying exposure and applying the patch to prevent potential code execution and data exposure.

• Potential code execution with runner process permissions
• Exposure of sensitive information
• Impact on availability due to command execution

Technical summary

The Omnigent framework does not properly validate agent bundles, allowing an authenticated user to execute a local command with runner process permissions. This can lead to exposure of sensitive information and impact availability. The vulnerability is caused by the validate_agent_bundle function in omnigent/server/bundles.py not rejecting a tools..callable dotted Python path. The _resolve_spec_callable function in omnigent/runner/tool_dispatch.py imports the specified module, and _execute_spec_callable_tool invokes the resolved function, allowing a bundle to select subprocess.check_output and execute a local command with the runner process permissions.

Defensive priority

Defenders should prioritize verifying exposure and applying the patch to prevent potential code execution and data exposure.

Recommended defensive actions

• Verify exposure by checking if the Omnigent framework version is prior to 0.3.0
• Apply the patch by updating to version 0.3.0 or later
• Monitor for suspicious activity and potential code execution attempts
• Review compensating controls for exposed systems while remediation is scheduled and verified
• Check relevant monitoring, detection, and logs for exposed assets that need extra review
• Track exceptions, retest remediated assets, and close the item only after evidence is documented
• Confirm whether affected product deployments exist in managed environments and assign an owner for follow-up

Evidence notes

The vulnerability is caused by the validate_agent_bundle function in omnigent/server/bundles.py not rejecting a tools..callable dotted Python path. The _resolve_spec_callable function in omnigent/runner/tool_dispatch.py imports the specified module, and _execute_spec_callable_tool invokes the resolved function.

Sources and references

Methodology and review provenance

AI-assisted synthesis based on stored public vulnerability evidence. System validation, approval state, and publication status do not by themselves establish human review of this revision. PatchSiren helps prioritize defensive review and does not prove exposure or remediation on any system.