mcp-developer
Use when building, debugging, or extending MCP servers or clients that connect AI systems with external tools and data sources. Invoke to implement tool handlers, configure resource providers, set up stdio/HTTP/SSE transport layers, validate schemas with Zod or Pydantic, debug protocol compliance issues, or scaffold complete MCP server/client projects using TypeScript or Python SDKs.
90
100%
Does it follow best practices?
Impact
81%
1.15xAverage score across 6 eval scenarios
Advisory
Suggest reviewing before use
Evaluation results
97%
56%Inventory Analytics MCP Server
Python MCP server with Pydantic validation
Correct Python packages
50%
100%
Pydantic input models
0%
100%
Pydantic field constraints
0%
100%
Async handler functions
0%
100%
Python decorator pattern
70%
100%
Stderr logging
100%
100%
McpError wrapping
0%
100%
Context manager cleanup
0%
62%
stdio_server transport
62%
100%
Tool inputSchema defined
50%
100%
No hardcoded credentials
100%
100%
Design explanation
100%
100%
100%
20%Developer Documentation MCP Server
TypeScript MCP resource provider
Correct npm package
100%
100%
Zod for validation
0%
100%
Server capabilities declared
100%
100%
Typed request schema constants
100%
100%
Hierarchical resource URIs
100%
100%
MIME types on resources
100%
100%
Stderr for logging
100%
100%
McpError with ErrorCode
0%
100%
StdioServerTransport
100%
100%
Resource not found handling
100%
100%
Design explanation
100%
100%
TypeScript configuration
100%
100%
76%
19%Public Data API MCP Gateway
MCP server rate limiting and timeouts
Rate limiting logic
100%
100%
Rate limit error code
0%
0%
Request timeout
58%
66%
Timeout error handling
25%
100%
Stateless tool design
87%
100%
Idempotency support
12%
0%
Input validation
25%
50%
McpError wrapping
0%
100%
Configuration file
87%
100%
Design explanation
100%
100%
No hardcoded secrets
100%
100%
MCP implementation file
100%
100%
62%
-30%Recipe Assistant MCP Server
TypeScript McpServer API with prompt templates
McpServer class used
100%
0%
server.tool() registration
100%
0%
Inline Zod validation
100%
0%
server.resource() registration
100%
0%
Prompt templates implemented
100%
100%
Prompt returns PromptMessage array
100%
100%
Stderr logging
100%
100%
McpError with ErrorCode
0%
100%
StdioServerTransport
100%
100%
Schema definitions file
100%
100%
Configuration file
100%
100%
Design explanation
100%
100%
60%
-2%Project Documentation MCP Server
Python FastMCP with resource templates
FastMCP class used
100%
0%
@mcp.tool() decorator
100%
0%
@mcp.resource() decorator
100%
0%
Resource template URIs
100%
66%
Pydantic BaseModel for tools
0%
100%
Field constraints present
0%
100%
Stderr logging
0%
100%
McpError error wrapping
0%
100%
mcp.run() entrypoint
100%
0%
Async handler functions
0%
100%
Schema definitions file
100%
100%
Configuration file
100%
100%
Design explanation
100%
100%
94%
6%Secure Workspace File Assistant MCP Server
MCP server security and authentication
No hardcoded API key
100%
100%
Environment variable documented
100%
100%
Sensitive data not in resources
100%
100%
Path traversal prevention
100%
100%
Path validation error is McpError
0%
0%
Input validation present
75%
100%
Descriptive tool names
66%
100%
Descriptive tool descriptions
100%
100%
Stderr-only logging
66%
100%
Schema definitions file
100%
100%
Configuration file
100%
100%
Design explanation
100%
100%
- Repository
- jeffallan/claude-skills
- Commit
5b76101
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.