personalization

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation provides instructions for setting up the environment using npx -y @coldiq/mcp@latest, which installs the author's Model Context Protocol (MCP) server package.
  • [COMMAND_EXECUTION]: The skill utilizes specialized MCP tools (e.g., mcp__coldiq__extract_post_engagement, mcp__coldiq__find_signals) to perform data enrichment and verification tasks.
  • [PROMPT_INJECTION]: The skill is designed to ingest and summarize external data from LinkedIn and company websites (Data Buckets 1-6).
  • Ingestion points: Social media content and company news fetched via mcp__coldiq__fetch_page_content and engagement tools.
  • Boundary markers: None explicitly defined in the instructions.
  • Capability inventory: Tools are limited to data enrichment, social intelligence, and email validation.
  • Sanitization: No explicit sanitization or filtering logic is provided for the fetched data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 06:37 PM
Security Audit — agent-trust-hub — personalization