We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/AI-Riksarkivet/oxenstierna'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
Makefile•1.29 KiB
.PHONY: install build dev serve serve-stdio inspect inspect-verbose clean all
# Install all dependencies
install:
npm install
uv sync
# Build the frontend
build:
npm run build
# Watch mode for frontend development
dev:
npm run dev
# Run server in HTTP mode (for testing with basic-host)
serve: build
@lsof -ti:3001 | xargs kill -9 2>/dev/null || true
uv run python -m ra_mcp_viewer_mcp.server
# Run server in stdio mode (for Claude Desktop)
serve-stdio: build
uv run python -m ra_mcp_viewer_mcp.server --stdio
# Open MCPJam inspector (run 'make serve' first in another terminal)
inspect:
@echo "Opening inspector for http://localhost:3001/mcp"
@echo "Note: Run 'make serve' in another terminal first"
npx @mcpjam/inspector@latest --url http://localhost:3001/mcp
# Open MCPJam inspector with verbose logging
inspect-verbose:
@echo "Opening inspector for http://localhost:3001/mcp (verbose)"
@echo "Note: Run 'make serve' in another terminal first"
npx @mcpjam/inspector@latest -v --url http://localhost:3001/mcp
# Clean build artifacts
clean:
rm -rf src/ra_mcp_viewer_mcp/dist node_modules .venv
# Full setup: install and build
all: install build
# Expose local server to the internet using Cloudflare Tunnel, than add: <tunnel_url>/mcp
tunnel:
npx cloudflared tunnel --url http://localhost:3001