PatchSiren

PatchSiren cyber security CVE debrief

CVE-2026-109437 CognicellAI CVE debrief

CVE-2026-109437 debrief: Cognition before 0.13.0 has a critical unauthenticated remote code execution vulnerability due to its agent sandbox backend defaulting to local execution and lacking REST/SSE API authentication. This vulnerability allows attackers to create sessions and post messages without credentials, leading to potential remote code execution as the server process. Defenders should assess exposure and apply remediation to prevent potential attacks. The CVE record and NVD entry provide details, but verification from official sources is necessary.

Vendor
CognicellAI
Product
Cognition
CVSS
CRITICAL 9.2
CISA KEV
Not listed in stored evidence
Original CVE published
2026-10-19
Original CVE updated
2026-10-20
Advisory published
2026-10-19
Advisory updated
2026-10-20

Who should care

Defenders managing Cognition deployments, especially those using versions before 0.13.0, should assess exposure and apply remediation to prevent potential remote code execution. Defenders should verify Cognition deployment versions and configurations to identify potential exposure. Security teams should review and restrict access to the REST/SSE API and implement authentication for the agent sandbox backend.

Why it matters

CVE-2026-109437 is a critical unauthenticated remote code execution vulnerability in Cognition before 0.13.0. Defenders should verify exposure and apply remediation to prevent potential attacks.

  • Remote code execution as the server process
  • Unauthenticated access to the REST/SSE API
  • Potential for attackers to create sessions and post messages without credentials
  • Verification of Cognition deployment versions and configurations

Technical summary

Cognition before 0.13.0 defaults its agent sandbox backend to local execution and lacks authentication on its REST/SSE API, allowing attackers to create sessions and post messages without credentials, leading to remote code execution as the server process. The vulnerability affects Cognition versions before 0.13.0, and defenders should verify exposure and apply remediation. The REST/SSE API lacks authentication, allowing attackers to create sessions and post messages without credentials. The agent sandbox backend defaults to local execution, which can lead to remote code execution as the server process.

Defensive priority

Defenders should prioritize verifying exposure and applying remediation for Cognition deployments, especially those using versions before 0.13.0.

Recommended defensive actions

  • Verify Cognition deployment versions and configurations to identify potential exposure.
  • Apply remediation by updating to Cognition version 0.13.0 or later.
  • Review and restrict access to the REST/SSE API.
  • Implement authentication for the agent sandbox backend.
  • 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.

Evidence notes

The CVE record and NVD entry provide details on the vulnerability, but its scope and remediation require verification from official sources. The vulnerability affects Cognition versions before 0.13.0, and defenders should verify exposure and apply remediation. The REST/SSE API lacks authentication, allowing attackers to create sessions and post messages without credentials. The agent sandbox backend defaults to local execution, which can lead to remote code execution as the server process.

Sources and references

Verified primary and authoritative sources

  • CVE-2026-109437 CVE Program record

    Publisher, destination, and source semantics verified

    URL: https://www.cve.org/CVERecord?id=CVE-2026-109437

    CVE Program - Official CVE Program record with source-provided CVE metadata.

  • CVE-2026-109437 NVD vulnerability detail

    Publisher, destination, and source semantics verified

    URL: https://nvd.nist.gov/vuln/detail/CVE-2026-109437

    NIST National Vulnerability Database - Official NIST NVD detail page and source-specific vulnerability assessment.

Supplemental 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.