personalization

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install and run the @coldiq/mcp package using npx. This package is provided by the skill's author (ColdIQ) and is a standard way to extend the agent's capabilities with specialized data sourcing tools.
  • [COMMAND_EXECUTION]: The documentation includes a command to initiate an MCP server (npx -y @coldiq/mcp@latest). This is intended for environment setup to allow the agent to interface with the vendor's API for data enrichment.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and summarize external content such as LinkedIn posts and company news (via mcp__coldiq__extract_post_engagement and mcp__coldiq__fetch_page_content) to generate email personalization. This represents a potential surface for indirect prompt injection if the external source contains malicious instructions.
  • Ingestion points: External data is ingested through the ColdIQ MCP tools (mcp__coldiq__extract_post_engagement, mcp__coldiq__fetch_page_content) referenced in SKILL.md.
  • Boundary markers: The instructions do not define specific boundary markers or delimiters for the ingested data.
  • Capability inventory: The skill uses ingested data to generate AI prompts and content for email campaigns.
  • Sanitization: There are no explicit instructions for sanitizing or filtering the content retrieved from external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 08:55 AM
Security Audit — agent-trust-hub — personalization