ClawHub
Back to skill

Security audit

Cmclaw

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed connector to YouCloud Creative Manager for ad-strategy analysis, with expected credentialed API calls and no hidden local persistence or destructive behavior.

Use this skill only if you have a YouCloud Creative Manager account and are comfortable sending the relevant prompt, selected material scope, and account-derived identifiers to the YouCloud API using your YOUCLOUD_API_KEY. Avoid setting DAM_API_BASE unless you trust the endpoint.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill declares access to environment variables and makes authenticated network calls, but does not explicitly declare permissions beyond metadata hints. That weakens least-privilege review and can cause platforms or reviewers to miss that the skill can exfiltrate secrets or send user content to external services. In this context, the skill sends prompts and account-derived identifiers to a remote API, so undisclosed network/env capabilities are materially relevant.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger conditions are broad and ambiguous, covering generic requests like ad strategy analysis, inspiration, or mention of common phrases. This can cause the skill to activate unexpectedly on unrelated conversations, leading to unnecessary transmission of user prompts to a third-party backend and confusing or privacy-impacting behavior. Because the skill also performs multi-step prompting and remote API calls, accidental activation increases exposure.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.