keyapi-twitter-content-analytics

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and ingests public Twitter/X user-generated content via the KeyAPI MCP server (https://mcp.keyapi.ai/twitter/mcp) — see SKILL.md workflow nodes like get_single_tweet_data, get_comments, search and scripts/run.js (callTool/convertImages) — and then the agent reads and synthesizes that content to decide follow-up analyses and API calls, so untrusted third-party text can materially influence behavior.