The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected. The skill is designed as an educational guide for developers.
[COMMAND_EXECUTION]: The documentation includes standard development commands for project initialization and debugging, such as npx @modelcontextprotocol/create-server and npx @modelcontextprotocol/inspector. These are official utilities provided for MCP development.
[EXTERNAL_DOWNLOADS]: The skill references standard package installations from official registries (NPM and PyPI). These include well-known libraries such as @modelcontextprotocol/sdk, mcp, zod, and pydantic, which are essential for the protocol implementation described.
[INDIRECT_PROMPT_INJECTION]: While the skill describes how to build servers that ingest external data (via tool arguments and resource URIs), it provides explicit and robust security patterns to mitigate injection risks. These include mandatory schema validation using Zod and Pydantic, regular expression checks for SQL injection prevention, and path resolution logic to prevent directory traversal attacks.

mcp-developer