ClawHub

Mistral AI

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Mistral AI connector that can read and change Mistral account data through OOMOL, with no evidence of hidden or malicious behavior.

Install only if you are comfortable connecting Mistral AI through OOMOL and letting the skill run oo connector actions on your behalf. Review exact payloads before uploads, sharing, updates, or deletes, and avoid exposing signed URLs, fileTransit download links, conversation history, or document text beyond the intended recipients.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The skill description says to use this skill for ANY Mistral AI request and instead of calling the API directly, which can cause the agent to invoke a highly privileged connector for broad, loosely matched tasks. That increases the chance of unintended activation and accidental execution of state-changing or destructive actions when a user only intended an informational or narrower operation.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The action description explicitly states it will download file contents and return an accessible download address, but it provides no warning about the sensitivity, sharing implications, or access-control risks of exposing downloaded content through fileTransit. In a connector skill that may be used broadly for reading and data handling, this omission can lead users or downstream agents to retrieve and redistribute sensitive files without recognizing the security consequences.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The action documentation states it returns a temporary signed URL to extracted document text but does not warn that the URL grants access to potentially sensitive document contents. In agent workflows, such URLs are often logged, copied into chats, or shared across systems, so lack of warning increases the chance of accidental disclosure even if the link is time-limited.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal