CVE-2026-37003

Received Received - Intake

Remote Code Execution in Agno up to 2.5.8 via Prompt Injection

Publication date: 2026-08-27

Last updated on: 2026-09-01

Assigner: MITRE

Description

Agno up to and including 2.5.8 is vulnerable to Remote Code Execution (RCE) via prompt injection. The PythonTools and ShellTools components pass unsanitized, LLM-generated arguments directly to execution sinks including exec(), runpy.run_path(), and subprocess.run(). An unauthenticated attacker can exploit this by embedding malicious instructions in content processed by the agent (such as web pages or documents), allowing for arbitrary code and OS command execution on the host server.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published

2026-08-27

Last Modified

2026-09-01

Generated

2026-09-17

AI Q&A

2026-08-27

EPSS Evaluated

2026-09-15

NVD

CVE-2026-37003

Affected Vendors & Products

Vendor	Product	Version / Range
agno	agno	to 2.5.8 (inc)

Helpful Resources

Exploitability

CWE

KEV

CWE ID	Description
CWE-94	The product constructs all or part of a code segment using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the syntax or behavior of the intended code segment.

Attack-Flow Graph

Executive Summary

Agno versions up to 2.5.8 are vulnerable to Remote Code Execution (RCE) through prompt injection. The PythonTools and ShellTools components execute unsanitized, LLM-generated arguments directly via functions like exec(), runpy.run_path(), and subprocess.run(). Attackers can embed malicious instructions in content processed by the agent, such as web pages or documents, to execute arbitrary code and OS commands on the server.

Detection Guidance

To detect this vulnerability, monitor for suspicious subprocess or exec() calls originating from Agno processes. Check logs for unexpected command executions or file modifications. Use tools like ps, netstat, or strace to inspect running processes and system calls. Look for unauthorized network connections initiated by Agno or its child processes.

Impact Analysis

This vulnerability allows unauthenticated attackers to execute arbitrary code and OS commands on the host server. This could lead to full system compromise, data theft, unauthorized access, or disruption of services. Systems using Agno for processing untrusted content are at high risk.

Compliance Impact

This RCE vulnerability could lead to unauthorized access or data breaches, violating GDPR and HIPAA requirements for data protection and access control. Organizations may face legal penalties, reputational damage, and loss of compliance certifications if exploited.

Mitigation Strategies

Immediately upgrade Agno to a version beyond 2.5.8 to remove the vulnerable components. Disable or restrict access to the PythonTools and ShellTools components if they are not essential. Review and sanitize all inputs processed by the agent to prevent prompt injection attacks.

Hi! I’m here to help you understand CVE-2026-37003. Ask me anything about the vulnerability, its impact, or mitigation strategies.

0/70