ClawHub
Back to skill

Security audit

crm-enrichment

Security checks across malware telemetry and agentic risk

Overview

The skill mostly matches CRM enrichment, but it adds unrelated GitHub account-starring and Cargo session-attribution behavior that users should review before installing.

Install only if you are comfortable using Cargo for CRM enrichment, sending the relevant CRM identifiers to that provider, and manually approving any paid or full-list run. Do not let the skill star the GitHub repository unless you deliberately want that public account action, and consider skipping the session-attribution command if you do not want extra usage metadata recorded.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (3)

Description-Behavior Mismatch

Medium
Confidence
98% confidence
Finding
The skill includes a post-task GitHub starring workflow that is unrelated to CRM enrichment and introduces an unnecessary side effect on the user's external account. Even though it asks for consent first, embedding account-modifying promotional behavior inside an operational skill violates least privilege and creates pressure to perform an action outside the stated purpose of the skill.

Context-Inappropriate Capability

High
Confidence
99% confidence
Finding
The `gh api -X PUT /user/starred/getcargohq/gtm-skills` command directly modifies the user's GitHub account but is not required for CRM enrichment. Any skill that includes commands to mutate third-party accounts outside its business function creates an opportunity for unauthorized or socially engineered actions, especially when bundled with legitimate operational steps.

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The session upsert block performs local plugin inspection and workspace/session tracking that is not necessary to enrich CRM records. This creates undisclosed telemetry and persistent metadata about the user's session, which expands data collection beyond the skill's stated purpose and could leak usage patterns or identifiers to the provider.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.