linkedin-hooks

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to download the vendor's official MCP package using npx -y @coldiq/mcp@latest from the NPM registry to enable data-driven features.
  • [REMOTE_CODE_EXECUTION]: To utilize the recommended data pulling capabilities, the skill provides a command to execute the latest @coldiq/mcp package directly from the remote NPM registry.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to ingest and process untrusted data from external social media sources.
  • Ingestion points: Data is pulled from the environment through tools including mcp__coldiq__extract_post_engagement, mcp__coldiq__find_influencers, and mcp__coldiq__find_signals.
  • Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat the retrieved social media data as untrusted or to ignore any embedded instructions within that data.
  • Capability inventory: The skill uses the ingested data to generate human-readable LinkedIn hooks and post headlines.
  • Sanitization: No sanitization, escaping, or validation steps are documented for the external content before it is processed by the AI.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 06:37 PM
Security Audit — agent-trust-hub — linkedin-hooks