The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill initiates the installation of the mcporter package from the official Node.js package registry (npm).
Evidence: The YAML frontmatter specifies an installation step for the mcporter package using the node package manager.
[COMMAND_EXECUTION]: The CLI tool provides a mechanism to execute arbitrary local shell commands when configured to use stdio-based MCP servers.
Evidence: The documentation provides an example command: mcporter call --stdio "bun run ./server.ts" scrape url=https://example.com, which directly executes local scripts.
[INDIRECT_PROMPT_INJECTION]: As the skill is designed to ingest and process outputs from external MCP servers, it is susceptible to indirect prompt injection if those servers return malicious instructions.
Ingestion points: Data returned from local stdio tools or remote HTTP tools via mcporter call (SKILL.md).
Boundary markers: No specific delimiters or "ignore instructions" warnings are defined in the skill instructions.
Capability inventory: Shell execution (via stdio), network communication (via HTTP), and file configuration management (SKILL.md).
Sanitization: The documentation does not specify sanitization or validation of the content returned by tools before it is processed by the agent.
[DYNAMIC_EXECUTION]: The skill includes functionality to generate new CLI tools and TypeScript client/type definitions based on remote server schemas.
Evidence: Commands such as mcporter generate-cli and mcporter emit-ts facilitate the creation of executable code or configurations at runtime.

mcporter