CVE-2026-51904: Superagi SuperAGI vulnerability

Q: Who can exploit this issue?

An attacker must be remotely authenticated as a user in one organization. They can target agents owned by a different organization if they can supply that agent's identifier to the affected endpoints.

Q: Which actions and endpoints are affected?

The affected controller functions are create_agent_execution and create_agent_run. They are reached through /agentexecutions/add and /agentexecutions/add_run, allowing creation or initiation of execution records for cross-organization agents.

Q: Are all deployments affected by default?

The issue is described in affected SuperAGI source snapshots up to v0.0.14. The provided information does not establish whether a particular deployment's default configuration exposes these endpoints or includes the affected code.

Q: How can I determine whether my instance is affected?

Review superagi/controllers/agent_execution.py and check whether create_agent_execution and create_agent_run verify that the caller's organization owns the caller-supplied agent_id before creating or starting an execution. Also assess whether authenticated users can access /agentexecutions/add or /agentexecutions/add_run.

Published Oct 2, 2026

Updated

SuperAGI up to v0.0.14 contains an improper access control vulnerability in the agent execution controller. In affected source snapshots, createagentexecution and createagentrun in superagi/controllers/agentexecution.py accept a caller-supplied agentid and fail to verify that the referenced agent belongs to the authenticated user's organization. A remote authenticated attacker from one organization can create or start execution records for agents owned by another organization through /agentexecutions/add or /agentexecutions/addrun.

Affected Software

1 affected component

superagi SuperAGI<=0.0.14

Event History

Oct 2, 2026

CVE Published

via MITRE·12:00 AM

Data Sourced

via MITRE·12:00 AM

Description

Data Sourced

via NVD·04:16 PM

Description

Frequently Asked Questions

Who can exploit this issue?

An attacker must be remotely authenticated as a user in one organization. They can target agents owned by a different organization if they can supply that agent's identifier to the affected endpoints.

Which actions and endpoints are affected?

The affected controller functions are create_agent_execution and create_agent_run. They are reached through /agentexecutions/add and /agentexecutions/add_run, allowing creation or initiation of execution records for cross-organization agents.

Are all deployments affected by default?

The issue is described in affected SuperAGI source snapshots up to v0.0.14. The provided information does not establish whether a particular deployment's default configuration exposes these endpoints or includes the affected code.

How can I determine whether my instance is affected?

Review superagi/controllers/agent_execution.py and check whether create_agent_execution and create_agent_run verify that the caller's organization owns the caller-supplied agent_id before creating or starting an execution. Also assess whether authenticated users can access /agentexecutions/add or /agentexecutions/add_run.

CVE-2026-51904: Superagi SuperAGI vulnerability

Affected Software

Event History

Frequently Asked Questions

Who can exploit this issue?

Which actions and endpoints are affected?

Are all deployments affected by default?

How can I determine whether my instance is affected?

Company

Resources

Contact

Frequently Asked Questions

Who can exploit this issue?

Which actions and endpoints are affected?

Are all deployments affected by default?

How can I determine whether my instance is affected?