linkedin-campaign-complete

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The skill’s workflow instructs first-party enrichment via ColdIQ MCP/API (e.g., mcp__coldiq__find_signals, mcp__coldiq__extract_post_engagement, mcp__coldiq__fetch_page_content) and only falls back to a “manual tier list” when ColdIQ isn’t available, with no runtime path described that directly ingests arbitrary outsider-authored free text from an inbox/feed/queue.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill makes runtime calls to ColdIQ to fetch personalization data that is injected into prompts (see https://api.coldiq.com) and also recommends running npx -y @coldiq/mcp@latest, which fetches and executes remote code, so the skill relies on external content that controls agent behavior.