twitter-automation

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

• Third-party content exposure detected (high risk: 0.75). The required workflow is driven by Twitter toolkit tool calls (e.g., TWITTER_RECENT_SEARCH / TWITTER_FULL_ARCHIVE_SEARCH) that return tweet/user content at runtime; that content is outsider-authored free text (public tweets) and would be ingested into the agent’s LLM context via the tool response payload.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires adding the external MCP endpoint https://rube.app/mcp which the agent will contact at runtime (e.g., via RUBE_SEARCH_TOOLS) to fetch tool schemas/instructions that directly control agent behavior.