The Agent Skills Directory

[SAFE]: The skill provides instructional content and code templates for building MCP servers and clients. The primary purpose is developer enablement.
[EXTERNAL_DOWNLOADS]: The skill references official and well-known developer tools and libraries including @modelcontextprotocol/sdk (NPM), mcp (PyPI), zod, and pydantic. These are standard dependencies for the Model Context Protocol ecosystem.
[COMMAND_EXECUTION]: Code examples include standard development commands like npx @modelcontextprotocol/create-server and pip install mcp. These are used for project initialization and dependency management in a way consistent with the skill's stated purpose.
[CREDENTIALS_UNSAFE]: The skill includes explicit security constraints in the SKILL.md file, specifically instructing the agent to never hardcode credentials or secrets and to return structured errors to avoid leaking system information.
[PROMPT_INJECTION]: No evidence of malicious instructions or behavior-altering patterns was found. The skill uses instructional language to define a "specialist" persona focused on implementation tasks.

mcp-developer