The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references documentation and setup files from the author's GitHub repository and provides instructions for running the omniroute package via npx.\n- [COMMAND_EXECUTION]: Provides configuration examples that execute the MCP server locally via npx and demonstrates API interactions using curl with shell variables.\n- [DYNAMIC_EXECUTION]: Includes tools that allow for the dynamic registration and invocation of additional skills at runtime (skill_register, skill_invoke).\n- [INDIRECT_PROMPT_INJECTION]:\n
Ingestion points: The agent processes data returned by 110 different tools and retrieves information through memory_search operations (SKILL.md).\n
Boundary markers: No explicit delimiters or instructions to disregard embedded commands are included in the skill definition.\n
Capability inventory: The skill possesses high-privilege capabilities, including memory modification (memory_add, memory_delete), skill lifecycle management, and administrative health checks for the underlying database (SKILL.md).\n
Sanitization: There is no evidence of sanitization or validation of data received from the external MCP server's tool outputs.

omni-mcp