CVE-2026-81096: Improper Control of Generation of Code ('Code Injection') in mims-harvard ToolUniverse
Description
ToolUniverse versions up to 1.2.6 contain a critical code injection vulnerability due to improper sandboxing of caller-supplied Python code. The server ran this code without authentication, allowing attackers to escape the sandbox and execute arbitrary code. Version 1.3.0 addresses these issues by adding authentication, restricting network binding, and improving attribute checks.
CVSS v4.0
Score 9.3critical
Affected software
mims-harvard
ToolUniverse
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
ToolUniverse's python_code_executor tool in versions up to 1.2.6 improperly controlled code generation by running caller-supplied Python inside a sandbox that could be escaped. The executor inspected source code against a deny list but left attribute-lookup builtins accessible, enabling attackers to traverse object hierarchies and import sensitive modules like process and subprocess. The server components bound to all interfaces with debugging enabled and no authentication, allowing unauthenticated remote code execution as the server process. Version 1.3.0 introduces bearer-token authentication, restricts binding to loopback, and hardens attribute inspections to mitigate this vulnerability.
Potential Impact
An unauthenticated attacker able to reach the server port could execute arbitrary code with the privileges of the server process. This allows full compromise of the host running ToolUniverse versions up to 1.2.6. The vulnerability is critical with a CVSS 4.0 score of 9.3, reflecting network attack vector, no required privileges or user interaction, and high impact on confidentiality, integrity, and availability.
Mitigation Recommendations
Version 1.3.0 of ToolUniverse includes fixes that add bearer-token authentication, restrict the server to bind only to the loopback interface by default, and improve attribute filtering to prevent sandbox escape. Users should upgrade to version 1.3.0 or later to remediate this vulnerability. No official patch or remediation level is explicitly stated beyond this version update. Until upgraded, restricting network access to the server ports and disabling debugging features may reduce exposure.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulnCheck
- Date Reserved
- 2026-08-26T16:00:32.344Z
- Cvss Version
- 4.0
- State
- PUBLISHED
Threat ID: 6a906bafacd9273b4983a0ef
Added to database: 08/27/2026, 16:54:07 UTC
Last enriched: 09/08/2026, 12:23:18 UTC
Last updated: 10/09/2026, 18:48:23 UTC
Views: 55
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.