Menu
Guide for developing MCP servers with Python SDK. Use when implementing MCP tools, resources, or prompts.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
A skill whose only resource is a binary asset for e2e binary-handling tests
A valid test skill with all features
Design RESTful APIs following best practices for consistency, usability, and scalability
Systematic code review methodology for identifying issues, suggesting improvements, and ensuring code quality
Write secure, efficient Dockerfiles following container best practices
Write clear, conventional commit messages that communicate changes effectively
| name | mcp-development |
| description | Guide for developing MCP servers with Python SDK. Use when implementing MCP tools, resources, or prompts. |
from mcp.server.fastmcp import FastMCP
mcp = FastMCP("skills-mcp")
@mcp.tool()
async def my_tool(param: str) -> str:
"""Tool description for LLM."""
return f"Result: {param}"
@mcp.resource("skills://{skill_name}")
async def get_skill(skill_name: str) -> str:
"""Return skill content."""
return skill_content
@mcp.prompt()
def review_skill(skill_name: str) -> str:
"""Create a prompt to review a skill."""
return f"Please review the skill: {skill_name}"
if __name__ == "__main__":
mcp.run(transport="streamable-http")
Use Context for logging and progress:
from mcp.server.fastmcp import Context
@mcp.tool()
async def long_task(ctx: Context) -> str:
await ctx.info("Starting task...")
await ctx.report_progress(50, 100)
return "Done"
See @src/skills_mcp/infrastructure/mcp/ for implementation examples.