mcp-server
Manages the setup and protocol handling for the deep-research MCP server, enabling external agents to query research pipelines.
Install
mkdir -p .claude/skills/mcp-server && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/19039" && unzip -o skill.zip -d .claude/skills/mcp-server && rm skill.zipInstalls to .claude/skills/mcp-server
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Load when working on the MCP server surface — the site exposed AS a tool other agents (Claude, Cursor, any MCP client) can call — or when touching src/mcp.js, the POST /mcp route, the deep_research tool, its input schema, the JSON-RPC 2.0 / Streamable-HTTP protocol handling (initialize / tools/list / tools/call / notifications/initialized), or debugging an MCP client that can't connect or call the tool. Covers the file-layout rule (pure protocol helpers static, pipeline dynamic-imported), how a tool call reuses chat.js's per-request setup (quota gate, model routing, usage/billing recording), how to add or change a tool, the shared seams (model-routing.js, billing.js), and the validation ladder (mcp.test.js → live JSON-RPC probe).Key capabilities
- →Expose deep-research pipeline as an MCP server
- →Handle JSON-RPC 2.0 over Streamable HTTP
- →Implement `initialize`, `tools/list`, `tools/call`, and `notifications/initialized` methods
- →Ensure MCP inherits site access control and quota gates
- →Record usage and billing for MCP calls
- →Add or change deep_research tool schema
How it works
The skill exposes the deep-research pipeline as an MCP server via a single POST to `/mcp`, handling JSON-RPC 2.0 requests to initialize, list tools, and call the `deep_research` tool, while ensuring access control and usage recording.
Inputs & outputs
When to use mcp-server
- →Debugging MCP tool calls
- →Adding new search tools
- →Managing JSON-RPC protocols
- →Integrating research agents
About mcp-server
Exposes the internal deep-research pipeline as an MCP server using JSON-RPC 2.0. It defines tool interfaces and transport helpers for inter-agent communication.
>-
When not to use it
- →When the pipeline should point inward instead of outward
- →When introducing model-driven tool selection on the inbound side
- →When adding a tool that is not a genuinely distinct outward capability
Limitations
- →The architecture argues MCP belongs on the outbound edge, NOT as internal plumbing
- →The module is split by import weight: pure protocol logic at top, pipeline behind dynamic import
- →Never introduce model-driven tool selection on the inbound side
How it compares
This skill provides a hand-rolled, minimal-dependency MCP server that exposes a deep-research pipeline as a tool for other agents, ensuring strict protocol adherence and shared access control, unlike a generic API endpoint.
Compared to similar skills
mcp-server side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| mcp-server (this skill) | 0 | 1mo | No flags | Advanced |
| mcp-builder | 136 | 4mo | Review | Advanced |
| mcp-integration | 21 | 9mo | Review | Intermediate |
| opencode-orchestrator-creator | 8 | 10mo | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
mcp-builder
anthropics
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).
mcp-integration
anthropics
This skill should be used when the user asks to "add MCP server", "integrate MCP", "configure MCP in plugin", "use .mcp.json", "set up Model Context Protocol", "connect external service", mentions "${CLAUDE_PLUGIN_ROOT} with MCP", or discusses MCP server types (SSE, stdio, HTTP, WebSocket). Provides comprehensive guidance for integrating Model Context Protocol servers into Claude Code plugins for external tool and service integration.
opencode-orchestrator-creator
IgorWarzocha
Creates universal OpenCode orchestrator folder structure with specialized agent that can manage swarm servers via curl commands
claude-opus-4-5-migration
anthropics
Migrate prompts and code from Claude Sonnet 4.0, Sonnet 4.5, or Opus 4.1 to Opus 4.5. Use when the user wants to update their codebase, prompts, or API calls to use Opus 4.5. Handles model string updates and prompt adjustments for known Opus 4.5 behavioral differences. Does NOT migrate Haiku 4.5.
mcp-management
mrgoonie
Manage Model Context Protocol (MCP) servers - discover, analyze, and execute tools/prompts/resources from configured MCP servers. Use when working with MCP integrations, need to discover available MCP capabilities, filter MCP tools for specific tasks, execute MCP tools programmatically, access MCP prompts/resources, or implement MCP client functionality. Supports intelligent tool selection, multi-server management, and context-efficient capability discovery.
n8n-mcp-orchestrator
manutej
Expert MCP (Model Context Protocol) orchestration with n8n workflow automation. Master bidirectional MCP integration, expose n8n workflows as AI agent tools, consume MCP servers in workflows, build agentic systems, orchestrate multi-agent workflows, and create production-ready AI-powered automation pipelines with Claude Code integration.