The Agent Skills Directory

[SAFE]: The skill serves as a reference for implementing Model Context Protocol (MCP) servers and FastAPI integrations. All dependencies mentioned (FastAPI, FastMCP, Pydantic, Uvicorn) are well-known and standard in the Python ecosystem.
[COMMAND_EXECUTION]: The skill provides instructions for running servers via the command line (e.g., using python -m and uvicorn). These are standard operations for the described development workflow.
[PROMPT_INJECTION]: The architecture describes ingestion points for external data that will be processed by an agent, which is a common surface for indirect prompt injection.
Ingestion points: The message argument in the send_to_agent tool and the MessageRequest Pydantic model in the FastAPI POST endpoint (found in SKILL.md).
Boundary markers: The provided snippets do not include explicit delimiters or instructions to ignore instructions within the message body.
Capability inventory: Messages are passed to agent.receive_message, which executes within the agent's logic context.
Sanitization: No input sanitization or validation is demonstrated in the boilerplate code.

mcp