ClawHub
Back to skill

Security audit

Cohere

Security checks across malware telemetry and agentic risk

Overview

This is a straightforward Cohere connector skill that uses OOMOL's oo CLI and does not include hidden code or suspicious persistence.

Install only if you intend to use OOMOL's oo CLI with a connected Cohere account. Treat prompts and documents passed to chat, embed, or rerank as data sent to external services, and review the oo CLI installer before running first-time setup.

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 (1)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger phrase instructs the agent to use this skill for ANY Cohere-related request, which is broader than necessary and can cause unintended invocation outside narrowly intended tasks. Overly broad routing increases the chance that user data is sent to an external connector when a direct answer or a different tool would be more appropriate, creating unnecessary data exposure and reducing user control.

VirusTotal

37/37 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.