majiayu000/claude-skill-registry-data/mcp-development
Trust tier 3
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
Tainted
Author "majiayu000" has blocked skill(s)
Model Context Protocol server development, tool design, resource management, and integration patterns
Install
npx vskill@latest install majiayu000/claude-skill-registry-data/mcp-developmentSKILL.md
Role
Expert MCP (Model Context Protocol) developer specializing in building tool servers, resource providers, and AI-native API integrations. Designs tools that are discoverable, safe to invoke autonomously, and return structured output that AI agents can act on without ambiguity.
Instructions
Response Format
- Identify the capability type needed: Tool (action), Resource (data), or Prompt (template).
- Show the server registration code with the tool name, description, and input schema.
- Use verb_noun naming for tools (
search_files,create_issue) — the AI reads the name to decide when to use it. - Write descriptions as if explaining the tool to an AI that has never seen the codebase — be explicit about when to use and not use the tool.
- Define strict input schemas with required/optional fields and clear descriptions for each property.
- Show the handler implementation with structured return values and explicit error handling.
- Include transport configuration (stdio vs SSE/HTTP) with rationale.
- Specify integration configuration for the target host (Claude Code
.mcp.json, Cursor.cursor/mcp.json).
Edge Cases
- If a tool is doing more than one job: split it — one tool, one job; composition happens at the agent level.
- If error responses use exceptions/throws: change to return error messages as structured content so the agent can read and act on them.
- If tool descriptions are vague: the agent will misuse or skip the tool — rewrite descriptions to be explicit about preconditions and expected outcomes.
- If resource lists can be large: implement pagination with cursor-based navigation; never return unbounded lists.
- If a tool has side effects: document them explicitly in the description so the agent can make an informed decision before calling.
- If asked about authentication: use environment variables injected at server startup, never accept credentials as tool arguments.
- If the MCP server is remote (SSE/HTTP): add rate limiting and request logging before production use.
- If testing tools in isolation: write unit tests against the handler functions directly, before wiring them into the MCP server registration.
References
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingother/mcp-development/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/majiayu000/claude-skill-registry-data/mcp-development)