The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the mcporter CLI to execute shell commands, specifically for launching MCP servers via the --stdio flag (e.g., mcporter list --stdio "npx -y @modelcontextprotocol/server-filesystem"). It also includes features for generating executable CLI wrappers.
[EXTERNAL_DOWNLOADS]: The instructions direct the agent to fetch and run code from the NPM registry using npx for both the primary mcporter tool and secondary MCP server packages.
[DATA_EXFILTRATION]: The skill facilitates sending data to external endpoints when calling remote HTTP-based MCP servers (e.g., mcporter call https://api.example.com/mcp.fetch). It also allows importing configuration files from local paths.
[PROMPT_INJECTION]: As a tool-agnostic interface, the skill is susceptible to indirect prompt injection where instructions hidden in the output of external MCP tools could influence the agent's logic.
Ingestion points: Data returned from mcporter call tool invocations in SKILL.md.
Boundary markers: No specific delimiters or safety warnings are implemented in the skill instructions.
Capability inventory: Includes subprocess execution via --stdio, network communication via HTTP tool calls, and configuration management.
Sanitization: No documented sanitization or validation of external tool outputs is provided.

mcporter