A local event gateway for AI code agents. Capture lifecycle events from Claude Code, Gemini CLI, and more — fan them out to any HTTP endpoint, webhook, or device.
<p align="center"> <img src="docs/banner.png" alt="agent-pulse" width="1000" /> </p> <h1 align="center">agent-pulse</h1> <p align="center"> <strong>A local event gateway for AI code agents.</strong><br/> Capture lifecycle events from Claude Code, Gemini CLI, and more — fan them out to any HTTP endpoint, webhook, or device. </p> <p align="center"> <a href="#installation">Install</a> • <a href="#quickstart">Quickstart</a> • <a href="#agent-setup">Agent Setup</a> • <a href="#configuration">Config</a> • <a href="#commands">Commands</a> </p> --- ## Why agent-pulse? Every AI code agent has its own hook system — different formats, different configs, scattered across projects. Adding a new webhook means editing every project's settings for every agent you use. **agent-pulse** solves this with a single global config: - **One config, all agents** — Define your clients once in `~/.config/agent-pulse/config.yaml`. Every project, every agent, every event routed through one place. - **Zero per-project setup** — No more duplicating hook configs across repositories. Register your hooks once and you're done. - **Fan-out to anything** — HTTP endpoints, webhooks, IoT devices (ESP32, Raspberry Pi, etc.), scripts — if it speaks HTTP, agent-pulse can reach it. - **Fully local** — Everything runs on your machine. No cloud, no accounts, no data leaves your network. - **Enrich with metadata** — Attach custom data to every lifecycle hook via a project-level `.agent-pulse.json` — project name, team, or anything your clients need. - **Filter what matters** — Route specific events to specific clients. Send `notification` events to your desk light, `session_start` to your dashboard, everything to your logger. - **Real-time streaming** — Connect to the SSE endpoint and watch events flow in real-time. Perfect for dashboards, browser UIs, or a simple `curl -N`. --- ## Diagram <p align="center"> <img src="docs/architecture.png" alt="Architecture diagram" width="1500" />
Google's AI-powered research notebook that ingests your documents and becomes an expert on your content. Generates audio overviews, study guides, FAQs, and interactive discussions from uploaded sources.
Google DeepMind's experimental AI agent that can navigate websites, fill forms, and complete multi-step browser tasks autonomously. Uses Gemini's multimodal understanding to interact with web interfaces.
Google DeepMind's universal AI assistant prototype that can see, hear, and respond in real-time through your device camera and microphone. Demonstrates the future of multimodal AI interaction.
Google Cloud's enterprise platform for building, deploying, and managing AI agents powered by Gemini. Supports multi-agent orchestration, tool integration, and enterprise governance.
Gemini's agentic research capability that autonomously browses the web, synthesizes information from dozens of sources, and produces comprehensive research reports on any topic.
Interactive coding and content creation agent that generates, previews, and iterates on code, documents, and interactive applications in a side panel. Supports HTML/CSS/JS, Python, and more.