x-scraper

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow reads outsider-authored free text from X itself by calling endpoints like GET /api/search with attacker-controlled query and returning matching posts/timelines (e.g., tweet text) to the agent (via twitter.fetcher.sh).

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly directs MCP-connected agents to fetch machine-readable/tool definitions at https://twitter.fetcher.sh/mcp (and related runtime artifacts like https://twitter.fetcher.sh/openapi.json and https://twitter.fetcher.sh/llms.txt / https://twitter.fetcher.sh/skill.md), which an agent can retrieve at runtime to discover tools and thereby directly influence its behavior.