I Can SEE EVERYTHING: Claude Code Hooks for Multi Agent Observability

Too Long; Didn't Watch — Summary

This video introduces a multi-agent observability system built with Claude Code Hooks, enabling real-time tracking and summarization of activities from numerous AI agents across different codebases and devices, crucial for scaling agentic coding workflows and maintaining control over complex AI systems.

Main Takeaways

• Scaling beyond a single AI agent creates a massive observability problem, making it difficult to track agent activities.
• Claude Code Hooks provide the foundation for building robust multi-agent observability systems by sending real-time events.
• A minimal client-server architecture (Bun server, SQLite, WebSockets) facilitates event collection, persistence, and real-time streaming to a UI.
• The system offers features like live activity pulse, event filtering, color-coded session tracking, and automatic summarization of events using fast, cost-effective LLMs like Haiku.
• Effective observability is paramount for steering, correcting, and controlling agentic systems, allowing engineers to scale their impact and move towards programmable agentic coding.

Detailed Summary

Introduction (0:00 - 0:48)

The video highlights the critical challenge of observability when scaling beyond a single Claude Code agent, emphasizing that multi-agent systems quickly become unmanageable without proper tracking. It introduces a solution using Claude Code Hooks to send events to a client-server architecture, enabling comprehensive visibility into the actions of multiple agents simultaneously.

Multi-Agent Observability System Overview (0:48 - 2:57)

This section details the multi-agent observability system, which ingests events from all running Claude Code instances, regardless of their tasks (subtasks, planning, building, web scraping, deep research, calling MCP servers). The architecture consists of multiple agents firing Claude Code Hooks, which hit a Bun server. This server stores data in an SQLite database for persistence and simultaneously streams events via WebSockets to a client application, ensuring real-time tracking across various devices and agents. The system provides notifications when agents require input or complete tasks, demonstrating its interactive capabilities.

Application UI/UX and Features (2:57 - 11:48)

The presenter showcases the application's user interface, which includes filters for different codebases (e.g., "infinite agentic loop" and "cloud code hook multi-agent observability"), session IDs, and event types. A live activity pulse visually represents the real-time event flow from all agents across applications. Key features include color-coding for applications and session IDs for easy identification, and automatic summarization of events (especially pre-tool and post-tool use) using a small, fast model like Haiku. This summarization provides quick, at-a-glance insights into agent activities, complementing the ability to view full event payloads and chat transcripts for detailed debugging, particularly with stop events.

Notable Quotes

"When it comes to multi-agent systems, observability is everything." — IndyDevDan

"With cloud code hooks sending events to a minimal client server architecture, you can see everything." — IndyDevDan

"This is the power of observability. As you scale up your agents, as you specialize and build agents into your workflows... you want to understand what they're doing." — IndyDevDan

"If you don't measure it, you can't improve it. If you don't monitor it, how will you know what's actually happening?" — IndyDevDan

"None of this generative AI technology matters if you have no idea what's going on." — IndyDevDan

"Prompting back and forth, one prompt at a time is not the way to engineer. It's a great place to start. It's a terrible place to finish." — IndyDevDan