systemprompt -core
The core MCP extension for Systemprompt MCP multimodal client
claude mcp add --transport stdio ejb503-systemprompt-mcp-core node build/index.js \ --env API_KEY="Your SystemPrompt API key"
How to use
SystemPrompt MCP Core is a server that manages AI agents, their system prompts, and tool configurations through the MCP protocol. It exposes capabilities for creating and configuring agents, managing prompts and resources, and integrating a customizable toolset that agents can use at runtime. The server also supports advanced features like sampling, real-time notifications, and API key authentication to enable secure, scalable usage with MCP-compatible clients. When used with the accompanying multimodal MCP client, you can drive voice-powered workflows, prompt-versioning, and resource management in an integrated workspace.
Once running, you can interact with the server’s agent tooling via MCP messages. Tools are organized into categories such as Prompt Management, Resource Management, System Tools, and Agent Management. For example, you can create or edit system prompts, manage agent configurations, list and read resources, and trigger system health checks. The sampling and notification features allow you to fine-tune response behavior and receive event-driven updates about agent activity. This makes it suitable for building extensible, versioned AI agent ecosystems that can be deployed across services and clients that support MCP.
How to install
Prerequisites:
- Node.js (recommended LTS, v14+)
- npm or pnpm
- Access to an API key from SystemPrompt (as described in the docs)
Installation steps:
-
Install dependencies: npm install
-
Build the server (transpile/compile TypeScript to JavaScript): npm run build
-
Run the server locally: node build/index.js
Optional (for development with auto-rebuild): npm run watch
If you prefer to install via Smithery (as documented in the project): npx -y @smithery/cli install systemprompt-agent-server --client claude
Configuration note:
- The MCP server is expected to be driven by a config object with mcpServers. For manual standalone usage, you can point a client to the built index.js file as shown in the README example.
Additional notes
Tips and caveats:
- You must supply an API key (API_KEY) to authorize requests to SystemPrompt services.
- The server uses MCP via stdio; debugging can be facilitated with the MCP Inspector tool.
- If you modify prompts or resources, consider versioning and clear metadata to keep agent configurations stable across deployments.
- Ensure build artifacts exist at build/index.js before starting the server; run npm run build to generate them.
- When integrating with clients, ensure MCP protocol compatibility, especially for sampling and notification features which depend on client support.
- If you encounter port or path issues, verify that the Node process has access to the path build/index.js and that environment variables are correctly passed to the runtime.
Related MCP Servers
mcp-graphql
Model Context Protocol server for GraphQL
mcp-google-map
A powerful Model Context Protocol (MCP) server providing comprehensive Google Maps API integration with LLM processing capabilities.
systemprompt-code-orchestrator
MCP server for orchestrating AI coding agents (Claude Code CLI & Gemini CLI). Features task management, process execution, Git integration, and dynamic resource discovery. Full TypeScript implementation with Docker support and Cloudflare Tunnel integration.
pluggedin-app
The Crossroads for AI Data Exchanges. A unified, self-hostable web interface for discovering, configuring, and managing Model Context Protocol (MCP) servers—bringing together AI tools, workspaces, prompts, and logs from multiple MCP sources (Claude, Cursor, etc.) under one roof.
rohlik
MCP server that lets you shop groceries across the Rohlik Group platforms (Rohlik.cz, Knuspr.de, Gurkerl.at, Kifli.hu, Sezamo.ro)
mcp -js
MCP server that exposes YepCode processes as callable tools for AI platforms. Securely connect AI assistants to your YepCode workflows, APIs, and automations.