The Agent Skills Directory

[SAFE]: The skill facilitates automation through a well-known third-party platform (Composio) and uses standard authentication and connection management workflows.
[EXTERNAL_DOWNLOADS]: The skill references external resources including the Composio documentation and the Rube MCP server endpoint. These are official service endpoints required for the skill's primary functionality.
[INDIRECT_PROMPT_INJECTION]: The skill employs dynamic tool discovery, fetching schemas from an external server at runtime. While this is a standard design for MCP-based integrations, it establishes an architectural ingestion point for external data.
Ingestion points: Tool schemas and execution plans are retrieved from the MCP server via RUBE_SEARCH_TOOLS and RUBE_GET_TOOL_SCHEMAS.
Boundary markers: The skill does not define specific delimiters for separating tool output from agent instructions.
Capability inventory: The skill utilizes RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH to perform actions based on discovered tool definitions.
Sanitization: No explicit validation or filtering of the dynamic schemas is present in the skill instructions.

pushover-automation