Live Exposure Defense: From CVE to Confirmed Exposure in 12 Hours – See more

New CVE Detected

CVE-2026-81096 – Unauthenticated RCE via Sandbox Escape – ToolUniverse ≤ 1.2.6

Be the first to know when new zero-days emerge:

Summary

CVE-2026-81096 is a critical unauthenticated remote code execution (RCE) vulnerability in ToolUniverse, an open-source Python framework (PyPI package tooluniverse) used to expose large numbers of AI/agent tools over HTTP and MCP (Model Context Protocol) servers. The flaw allows a remote, unauthenticated attacker to escape the sandbox protecting ToolUniverse’s built-in python_code_executor tool and execute arbitrary code as the server process. It is rated CRITICAL, with a CVSS v3.1 base score of 10.0 and a CVSS v4.0 score of 9.3.

Technical details

  • Root cause: The python_code_executor tool (implemented in python_executor_tool.py) enforces a denylist of dangerous attributes rather than an allowlist. The denylist fails to block dunder attributes reached via string-based attribute lookups or via modules that are otherwise permitted, letting an attacker "walk" from a literal object’s class to its base class and enumerate subclasses to reach the subprocess module and other dangerous primitives.
  • Compounding issue — import allowlist bypass: A per-call argument allows the import allowlist used by the sandbox to be expanded before the security checks that should constrain it are executed, further weakening the sandbox.
  • Compounding issue — missing authentication/network exposure: The bundled HTTP and MCP servers bind to all network interfaces (0.0.0.0) by default with debugging enabled and require no authentication, meaning the vulnerable code executor is reachable directly over the network without any credentials.
  • Attack vector: Network-based, no authentication or user interaction required (AV:N/AC:L/PR:N/UI:N). An attacker sends a crafted request/tool invocation containing attribute-lookup chains that reach subprocess (or similar) functionality through the exposed executor.
  • Impact: Full remote code execution as the ToolUniverse server process, resulting in complete loss of confidentiality, integrity, and availability on the host.

Affected software

  • ToolUniverse (PyPI package tooluniverse), all versions from initial release through 1.2.6
  • Fixed in version 1.3.0

Severity

  • CVSS v3.1 Base Score: 10.0 (Critical) — Vector: AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
  • CVSS v4.0 Score: 9.3 (Critical) — Vector: AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H

Mitigation and recommended actions

  • Immediate: Upgrade ToolUniverse to version 1.3.0 or later, which introduces bearer-token authentication for the HTTP/MCP servers, restricts the default bind address to loopback (localhost) instead of all interfaces, and hardens the attribute-validation logic used by the sandbox.
  • If immediate patching is not possible:
    • Do not expose ToolUniverse’s HTTP or MCP server ports to untrusted networks or the public internet; restrict access to 127.0.0.1/loopback or place them behind a properly authenticated reverse proxy/VPN.
    • Disable or remove the python_code_executor tool if it is not required for your deployment.
    • Disable debug mode on any exposed ToolUniverse server instances.
    • Monitor for unexpected outbound process execution or subprocess invocations originating from ToolUniverse server processes as an indicator of exploitation attempts.

References

Are you exposed?

Get a free report of your organization’s exposure to this CVE and threat

How IONIX’s External Exposure Management Platform Detects and Validates
Zero-Days to Shrink MTTR

1

Map your entire attack surface (continously)

IONIX uses multi-factor discovery methods, including DNS analysis, certificate mapping, metadata inspection, and more, to automatically map every internet-facing asset across your environment. This includes cloud instances, third-party platforms, shadow IT, and even forgotten infrastructure that traditional tools miss.

2

Monitor for new CVEs

Dozens of threat intel feeds using agentic technology are continuously analyzed to detect the appearance of proof-of-concept code, exploit kits, and indicators of active targeting. IONIX goes further by applying AI to proactively evaluate whether emerging vulnerabilities are likely to be exploited, even before PoCs go public.

3

Identify Potential External Exposures

Not all CVEs matter. IONIX filters vulnerabilities by asking attacker-centric questions: Can it be reached from the internet? Does it require authentication? Is it being exploited in the wild? This dramatically reduces noise and focuses teams on threats that can actually be weaponized.

4

Create Safe, Scalable Exploit Validations

IONIX transforms real-world PoCs into safe, non-intrusive test payloads that can be run in production environments without disruption. These simulations are precisely targeted to the systems that are vulnerable, ensuring rapid validation without unnecessary load.

5

Execute Exploit Validations

By combining context about software stack, versioning, exposure status, and reachability, IONIX ensures that only the right payloads are executed against the right assets, maximizing efficiency and minimizing risk.

6

Drive Fast and Actionable Remediation

Results are routed through integrations with ticketing, SOAR, and SIEM tools. Issues are written in plain language, bundled into remediation clusters, and prioritized based on asset criticality, exploitability, and blast radius. This shortens mean time to remediation (MTTR) and empowers teams to act with confidence.

Are you exposed?

Get a free report of your organization’s exposure to this CVE and threat

Subscribe to Threat Center RSS

Copy/paste the link below into your preferred RSS reader or follow these instructions to subscribe to Slack alerts.

Get Real-Time CVE Alerts to Your Email

Be the first to know when new zero-days emerge