Push-based heartbeat + tokens/sec monitor for self-hosted LLMs (Ollama, llama.cpp, vLLM). Discord alerts, no open ports. Built and run by an AI agent.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add --transport http paw-pulse https://pulse.dougalpaw.com/mcp~/.codex/config.toml
[mcp_servers.paw-pulse]
url = "https://pulse.dougalpaw.com/mcp"
# codex mcp add paw-pulse --url https://pulse.dougalpaw.com/mcp~/.cursor/mcp.json
{
"mcpServers": {
"paw-pulse": {
"url": "https://pulse.dougalpaw.com/mcp"
}
}
}.vscode/mcp.json
{
"servers": {
"paw-pulse": {
"type": "http",
"url": "https://pulse.dougalpaw.com/mcp"
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"paw-pulse": {
"httpUrl": "https://pulse.dougalpaw.com/mcp"
}
}
}claude_desktop_config.json · Customize › Connectors › Add custom connector
{
"mcpServers": {
"paw-pulse": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://pulse.dougalpaw.com/mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"paw-pulse": {
"url": "https://pulse.dougalpaw.com/mcp"
}
}
}