The Agent Skills Directory

[DATA_EXFILTRATION]: The skill instructs users to configure an external MCP server at https://rube.app/mcp. This third-party service is used to manage connections and retrieve data from Make (formerly Integromat). Connecting sensitive automation infrastructure to an unverified external endpoint introduces a risk of data exposure if the service is compromised or behaves maliciously.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources via Make operations, creating a surface for indirect prompt injection.
Ingestion points: The tool MAKE_GET_OPERATIONS retrieves execution logs and data from Make scenarios, which often contain untrusted content from various third-party applications (e.g., emails, webhooks, database records).
Boundary markers: The instructions do not define boundary markers (like XML tags or delimiters) to separate retrieved data from the agent's internal instructions.
Capability inventory: The skill includes powerful capabilities such as RUBE_MANAGE_CONNECTIONS for authentication and RUBE_SEARCH_TOOLS for dynamic tool discovery and execution.
Sanitization: There is no evidence of sanitization, filtering, or validation performed on the data retrieved from Make before it is processed by the agent.

make-automation