linkedin-ads
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs users to install and run the official ColdIQ MCP tool using
npx -y @coldiq/mcp@latest. This is a standard installation method for the author's own integration tools. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data such as competitor LinkedIn ad copy and headlines through the
mcp__coldiq__search_adstool. This presents a theoretical surface for indirect prompt injection if an attacker were to place malicious instructions within their ad copy. - Ingestion points: Output from competitive research tools (
mcp__coldiq__search_ads). - Boundary markers: None explicitly defined in the instructions for processing ad content.
- Capability inventory: File reading, network operations (via MCP), and prompt generation.
- Sanitization: Not explicitly documented; the skill relies on the agent's internal reasoning to analyze the gathered data.
Audit Metadata