The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/evaluation.py and scripts/connections.py scripts enable the execution of local shell commands to launch MCP servers via the stdio transport. This is a core feature for testing server implementations but allows for arbitrary command execution based on user-provided arguments.\n
Ingestion points: CLI arguments -c and -a passed to the evaluation script.\n
Capability: Spawns local subprocesses using the mcp library.\n- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch documentation and specifications from modelcontextprotocol.io and GitHub repositories belonging to the modelcontextprotocol organization. These are official and trusted sources for the protocol.\n- [PROMPT_INJECTION]: The evaluation framework processes external questions from XML files, which presents a surface for indirect prompt injection.\n
Ingestion points: scripts/evaluation.py reads task questions from an XML file provided by the user.\n
Boundary markers: Lacks specific delimiters or warnings to prevent the LLM from obeying instructions embedded in the evaluation questions.\n
Capability inventory: The script can execute local commands and perform network operations via the Anthropic API.\n
Sanitization: No validation or sanitization is performed on the questions before they are interpolated into the system prompt.

mcp-builder