mcp-builder

Name: mcp-builder
Author: ECNU-ICALK

524stars49forks

How to install mcp-builder?

ecnu-icalk/autoskill/mcp-builder

$npx -y skills add ECNU-ICALK/AutoSkill --skill mcp-builder

Installs into the current project.

Summary

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

Files · 1

View on GitHub

SKILL.md

1	# MCP Server Development Guide
2
3	## Overview
4
5	To create high-quality MCP (Model Context Protocol) servers that enable LLMs to effectively interact with external services, use this skill. An MCP server provides tools that allow LLMs to access external services and APIs. The quality of an MCP server is measured by how well it enables LLMs to accomplish real-world tasks using the tools provided.
6
7	---
8
9	# Process
10
11	## 🚀 High-Level Workflow
12
13	Creating a high-quality MCP server involves four main phases:
14
15	### Phase 1: Deep Research and Planning
16
17	#### 1.1 Understand Agent-Centric Design Principles
18
19	Before diving into implementation, understand how to design tools for AI agents by reviewing these principles:
20
21	Build for Workflows, Not Just API Endpoints:
22	- Don't simply wrap existing API endpoints - build thoughtful, high-impact workflow tools
23	- Consolidate related operations (e.g., `schedule_event` that both checks availability and creates event)
24	- Focus on tools that enable complete tasks, not just individual API calls
25	- Consider what workflows agents actually need to accomplish
26
27	Optimize for Limited Context:
28	- Agents have constrained context windows - make every token count
29	- Return high-signal information, not exhaustive data dumps
30	- Provide "concise" vs "detailed" response format options
31	- Default to human-readable identifiers over technical codes (names over IDs)
32	- Consider the agent's context budget as a scarce resource
33
34	Design Actionable Error Messages:
35	- Error messages should guide agents toward correct usage patterns
36	- Suggest specific next steps: "Try using filter='active_only' to reduce results"
37	- Make errors educational, not just diagnostic
38	- Help agents learn proper tool usage through clear feedback
39
40	Follow Natural Task Subdivisions:
41	- Tool names should reflect how humans think about tasks
42	- Group related tools with consistent prefixes for discoverability
43	- Design tools around natural workflows, not just API structure
44
45	Use Evaluation-Driven Development:
46	- Create realistic evaluation scenarios early
47	- Let agent feedback drive tool improvements
48	- Prototype quickly and iterate based on actual agent performance
49
50	#### 1.3 Study MCP Protocol Documentation
51
52	Fetch the latest MCP protocol documentation:
53
54	Use WebFetch to load: `https://modelcontextprotocol.io/llms-full.txt`
55
56	This comprehensive document contains the complete MCP specification and guidelines.
57
58	#### 1.4 Study Framework Documentation
59
60	Load and read the following reference files:
61
62	- MCP Best Practices: [📋 View Best Practices](./reference/mcp_best_practices.md) - Core guidelines for all MCP servers
63
64	For Python implementations, also load:
65	- Python SDK Documentation: Use WebFetch to load `https://raw.githubusercontent.com/modelcontextprotocol/python-sdk/main/README.md`
66	- [🐍 Python Implementation Guide](./reference/python_mcp_server.md) - Python-specific best practices and examples
67
68	For Node/TypeScript implementations, also load:
69	- TypeScript SDK Documentation: Use WebFetch to load `https://raw.githubusercontent.com/modelcontextprotocol/typescript-sdk/main/README.md`
70	- [⚡ TypeScript Implementation Guide](./reference/node_mcp_server.md) - Node/TypeScript-specific best practices and examples
71
72	#### 1.5 Exhaustively Study API Documentation
73
74	To integrate a service, read through ALL available API documentation:
75	- Official API reference documentation
76	- Authentication and authorization requirements
77	- Rate limiting and pagination patterns
78	- Error responses and status codes
79	- Available endpoints and their parameters
80	- Data models and schemas
81
82	To gather comprehensive information, use web search and the WebFetch tool as needed.
83
84	#### 1.6 Create a Comprehensive Implementation Plan
85
86	Based on your research, create a detailed plan that includes:
87
88	Tool Selection:
89	- List the most valuable endpoints/operations to implement
90	- Prioritize tools that enable the most common and important use cases
91	- Consider which tools work together to enable complex workflows
92
93	Shared Utilities and Helpers:
94	- Identify common API request patterns
95	- Plan pagination helpers
96	- Design filtering and formatting utilities
97	- Plan error handling strategies
98
99	Input/Output Design:
100	- Define input validation models (Pydantic for Python, Zod for TypeScript)
101	- Design consistent response formats (e.g., JSON or Markdown), and configurable levels of detail (e.g., Detailed or Concise)
102	- Plan for large-scale usage (thousands of users/resources)
103	- Implement character limits and truncation strategies (e.g., 25,000 tokens)
104
105	Error Handling Strategy:
106	- Plan g

People who liked this also viewed

6 picks

Type

mcp-builder

524stars49forks

View on GitHub

How to install mcp-builder?

ecnu-icalk/autoskill/mcp-builder

$npx -y skills add ECNU-ICALK/AutoSkill --skill mcp-builder

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# MCP Server Development Guide
2
3	## Overview
4
5	To create high-quality MCP (Model Context Protocol) servers that enable LLMs to effectively interact with external services, use this skill. An MCP server provides tools that allow LLMs to access external services and APIs. The quality of an MCP server is measured by how well it enables LLMs to accomplish real-world tasks using the tools provided.
6
7	---
8
9	# Process
10
11	## 🚀 High-Level Workflow
12
13	Creating a high-quality MCP server involves four main phases:
14
15	### Phase 1: Deep Research and Planning
16
17	#### 1.1 Understand Agent-Centric Design Principles
18
19	Before diving into implementation, understand how to design tools for AI agents by reviewing these principles:
20
21	Build for Workflows, Not Just API Endpoints:
22	- Don't simply wrap existing API endpoints - build thoughtful, high-impact workflow tools
23	- Consolidate related operations (e.g., `schedule_event` that both checks availability and creates event)
24	- Focus on tools that enable complete tasks, not just individual API calls
25	- Consider what workflows agents actually need to accomplish
26
27	Optimize for Limited Context:
28	- Agents have constrained context windows - make every token count
29	- Return high-signal information, not exhaustive data dumps
30	- Provide "concise" vs "detailed" response format options
31	- Default to human-readable identifiers over technical codes (names over IDs)
32	- Consider the agent's context budget as a scarce resource
33
34	Design Actionable Error Messages:
35	- Error messages should guide agents toward correct usage patterns
36	- Suggest specific next steps: "Try using filter='active_only' to reduce results"
37	- Make errors educational, not just diagnostic
38	- Help agents learn proper tool usage through clear feedback
39
40	Follow Natural Task Subdivisions:
41	- Tool names should reflect how humans think about tasks
42	- Group related tools with consistent prefixes for discoverability
43	- Design tools around natural workflows, not just API structure
44
45	Use Evaluation-Driven Development:
46	- Create realistic evaluation scenarios early
47	- Let agent feedback drive tool improvements
48	- Prototype quickly and iterate based on actual agent performance
49
50	#### 1.3 Study MCP Protocol Documentation
51
52	Fetch the latest MCP protocol documentation:
53
54	Use WebFetch to load: `https://modelcontextprotocol.io/llms-full.txt`
55
56	This comprehensive document contains the complete MCP specification and guidelines.
57
58	#### 1.4 Study Framework Documentation
59
60	Load and read the following reference files:
61
62	- MCP Best Practices: [📋 View Best Practices](./reference/mcp_best_practices.md) - Core guidelines for all MCP servers
63
64	For Python implementations, also load:
65	- Python SDK Documentation: Use WebFetch to load `https://raw.githubusercontent.com/modelcontextprotocol/python-sdk/main/README.md`
66	- [🐍 Python Implementation Guide](./reference/python_mcp_server.md) - Python-specific best practices and examples
67
68	For Node/TypeScript implementations, also load:
69	- TypeScript SDK Documentation: Use WebFetch to load `https://raw.githubusercontent.com/modelcontextprotocol/typescript-sdk/main/README.md`
70	- [⚡ TypeScript Implementation Guide](./reference/node_mcp_server.md) - Node/TypeScript-specific best practices and examples
71
72	#### 1.5 Exhaustively Study API Documentation
73
74	To integrate a service, read through ALL available API documentation:
75	- Official API reference documentation
76	- Authentication and authorization requirements
77	- Rate limiting and pagination patterns
78	- Error responses and status codes
79	- Available endpoints and their parameters
80	- Data models and schemas
81
82	To gather comprehensive information, use web search and the WebFetch tool as needed.
83
84	#### 1.6 Create a Comprehensive Implementation Plan
85
86	Based on your research, create a detailed plan that includes:
87
88	Tool Selection:
89	- List the most valuable endpoints/operations to implement
90	- Prioritize tools that enable the most common and important use cases
91	- Consider which tools work together to enable complex workflows
92
93	Shared Utilities and Helpers:
94	- Identify common API request patterns
95	- Plan pagination helpers
96	- Design filtering and formatting utilities
97	- Plan error handling strategies
98
99	Input/Output Design:
100	- Define input validation models (Pydantic for Python, Zod for TypeScript)
101	- Design consistent response formats (e.g., JSON or Markdown), and configurable levels of detail (e.g., Detailed or Concise)
102	- Plan for large-scale usage (thousands of users/resources)
103	- Implement character limits and truncation strategies (e.g., 25,000 tokens)
104
105	Error Handling Strategy:
106	- Plan g

People who liked this also viewed

6 picks

Type