promptz
MCP Server implementation for promptz.dev
claude mcp add --transport stdio cremich-promptz-mcp npx -y @promptz/mcp \ --env PROMPTZ_API_KEY="your-api-key-from-promptz.dev" \ --env PROMPTZ_API_URL="your-api-url-from-promptz.dev"
How to use
The promptz MCP server provides access to prompts from promptz.dev directly within Amazon Q Developer. It exposes two core capabilities: Prompts and Rules. Prompts are executable functions you can search for and run to insert or execute specific prompts in your workflow. Rules are executable functions that help you locate project rules and, when integrated with other tools, add or update them in your workspace. Once connected, you can interact with these capabilities using natural language, for example: “Search for CLI prompts about JavaScript” or “Find project rules for CDK Development.” The server authenticates with promptz.dev using your API credentials and communicates over HTTPS to fetch prompt data and project rules. The MCP client will invoke the appropriate function (prompt search, prompt execution, or rule search/update) based on your request, returning results or performing workspace updates as needed.
How to install
Prerequisites:
- Node.js (recommended) or access to a compatible MCP client (Amazon Q Developer)
- Access to promptz.dev MCP API credentials (API URL and API Key)
Option A: Quick start with npx (no local install)
- Ensure you have your promptz.dev API URL and API Key handy.
- In your MCP client settings, add a server definition using npx as shown below.
Code example (copy into your MCP client settings file, e.g., ~/.aws/amazonq/mcp.json):
{
"mcpServers": {
"promptz.dev": {
"command": "npx",
"args": ["-y", "@promptz/mcp"],
"env": {
"PROMPTZ_API_URL": "your-api-url-from-promptz.dev",
"PROMPTZ_API_KEY": "your-api-key-from-promptz.dev"
},
"disabled": false,
"autoApprove": []
}
}
}
Option B: Local installation (Node.js required)
- Clone the repository and install dependencies:
git clone https://github.com/cremich/promptz-mcp.git
cd promptz-mcp
npm install
- Build if required (check repository for build script):
npm run build
- Configure your MCP client to run the server locally via Node.js, for example:
{
"mcpServers": {
"promptz.dev": {
"command": "node",
"args": ["/path/to/promptz-mcp/build/index.js"],
"env": {
"PROMPTZ_API_URL": "your-api-url-from-promptz.dev",
"PROMPTZ_API_KEY": "your-api-key-from-promptz.dev"
},
"disabled": false,
"autoApprove": []
}
}
}
Troubleshooting tips:
- Verify API credentials are correct in the MCP client configuration.
- Ensure the server is reachable from the MCP client and that the correct command/args are configured.
- Check logs at ~/.promptz/logs/mcp-server.log for errors.
- Use the MCP Inspector feature to debug requests when available.
Additional notes
Tips and common considerations:
- The server exposes read access to prompts and rules; there are no write operations performed by the MCP server itself.
- Store API credentials securely in your MCP client configuration; avoid embedding secrets in shared code.
- Ensure HTTPS communication with promptz.dev API to protect credentials in transit.
- If prompts or rules are not returning expected results, verify that the API URL and API Key have the appropriate permissions for search and retrieval.
- When using npx, you don’t need to install the package locally; the MCP client will fetch the latest @promptz/mcp version on each run.
- For local development, use the inspector tool provided by the MCP framework to debug and test queries before using them in your workflow.
Related MCP Servers
obsidian -tools
Add Obsidian integrations like semantic search and custom Templater prompts to Claude or any MCP client.
mcp
Octopus Deploy Official MCP Server
furi
CLI & API for MCP management
mcp -arangodb
This is a TypeScript-based MCP server that provides database interaction capabilities through ArangoDB. It implements core database operations and allows seamless integration with ArangoDB through MCP tools. You can use it wih Claude app and also extension for VSCode that works with mcp like Cline!
aws
An MCP(Model Context Protocol) Server for AWS services
CodeRAG
Advanced graph-based code analysis for AI-assisted software development