The Agent Skills Directory

[SAFE]: The skill consists of documentation and configuration instructions for the Superset MCP service using standard CLI commands for various agent platforms.
[INDIRECT_PROMPT_INJECTION]: The skill manages external coding agents and Git repositories, creating a surface where untrusted data (repository content or agent outputs) could influence the host agent. * Ingestion points: External Git repositories and responses from parallel coding agents accessed via the Superset API. * Boundary markers: The skill documentation notes that sensitive operations are marked with protocol hints (destructiveHint) to prompt for user confirmation. * Capability inventory: Includes tools for launching coding agent sessions, creating workspaces, and opening terminal PTYs in isolated environments (SKILL.md). * Sanitization: No specific sanitization or filtering logic is described in the skill configuration.

superset-mcp