The Agent Skills Directory

Prompt Injection (LOW): The skill is susceptible to indirect prompt injection because it processes untrusted data from the Close CRM API. Ingestion points: Data enters the agent context through SMS message content (CLOSE_CREATE_SMS), lead notes (CLOSE_GET_NOTE), and custom field values (CLOSE_CREATE_LEAD). Boundary markers: Absent. The instructions do not specify any delimiters or warnings to the agent regarding embedded instructions in the CRM data. Capability inventory: The skill utilizes tools with write and delete capabilities in the CRM environment, such as CLOSE_CREATE_LEAD, CLOSE_CREATE_CALL, and CLOSE_DELETE_CALL. Sanitization: Absent. No input validation or escaping logic is defined for the external data being processed.
External Downloads (LOW): The skill requires connecting to an external MCP server (https://rube.app/mcp) which is not on the list of trusted providers.
No Code (SAFE): The skill does not provide any executable code or scripts, reducing the direct attack surface.

close-automation