ClawHub
Back to skill

Security audit

Evervault

Security checks for vulnerabilities and agentic risk

Overview

The skill is a mostly coherent Evervault connector, but it gives broad automatic guidance for sensitive encryption and decryption actions without clear user confirmation or scoping.

Review this skill carefully before installing. It appears designed to use OOMOL's oo CLI for Evervault, but only invoke it for explicit encrypt, decrypt, or token-inspection tasks, and confirm any decryption payload before running because decrypted output may reveal sensitive data.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The manifest and description claim this skill is for 'searching and reading data,' but the documented actions include `encrypt_json` and especially `decrypt_json`, which are not read-only operations and materially expand the skill's capabilities. This mismatch can mislead users or higher-level agents into invoking the skill under a weaker trust assumption, increasing the chance of unintended sensitive-data handling or disclosure.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger text says to use this skill for 'ANY Evervault request' and 'instead of calling the API directly,' which is overly broad and may cause the skill to be selected whenever Evervault is merely mentioned. In context, this is more dangerous because the skill exposes decryption functionality, so accidental invocation could route sensitive operations through a tool the user did not specifically request.

Static analysis

No suspicious patterns detected.