Hacker News Viewer

Show HN: Real-time dashboard for Claude Code agent teams

by simple10 on 4/1/2026, 4:24:28 PM

This project (Agents Observe) started as an exploration into building automation harnesses around claude code. I needed a way to see exactly what teams of agents were doing in realtime and to filter and search their output.<p>A few interesting learnings from building and using this:<p>- Claude code hooks are blocking - performance degrades rapidly if you have a lot of plugins that use hooks<p>- Hooks provide a lot more useful info than OTEL data<p>- Claude&#x27;s jsonl files provide the full picture<p>- Lifecycle management of MCP processes started by plugins is a bit kludgy at best<p>The biggest takeaway is how much of a difference it made in claude performance when I switched to background (fire and forget) hooks and removed all other plugins. It&#x27;s easy to forget how many claude plugins I&#x27;ve installed and how they effect performance.<p>The Agents Observe plugin uses docker to start the API and dashboard service. This is a pattern I&#x27;d love to see used more often for security (think Axios hack) reasons. The tricky bit was handling process management across multiple claude instances - the solution was to have the server track active connections then auto shut itself down when not in use. Then the plugin spins it back up when a new session is started.<p>This tool has been incredibly useful for my own daily workflow. Enjoy!

https://github.com/simple10/agents-observe

Comments

by: saadn92

The hooks performance finding matches what I&#x27;ve seen. I run multiple Claude Code agents in parallel on a remote VM and the first thing I learned was that anything blocking in the agent&#x27;s critical path kills throughput. Even a few hundred milliseconds per hook call compounds fast when you have agents making dozens of tool calls per minute.<p>The docker-based service pattern is smart too. I went a different direction for my own setup -- tmux sessions with worktree isolation per agent, which keeps things lightweight but means I have zero observability into what each agent is actually doing beyond tailing logs manually. This solves that gap in a way that doesn&#x27;t add overhead to the agent itself, which is the right tradeoff.<p>Curious about one thing -- how does the dashboard handle the case where a sub-agent spawns its own sub-agents? Does it track the full tree or just one level deep?

4/1/2026, 5:47:11 PM


by: petcat

Are you guys spending hundreds (or thousands) of dollars a day on Claude tokens? Holy crap. I can&#x27;t get more than one or two agents to do anything useful for very long before I&#x27;m hitting my usage limits.

4/1/2026, 5:34:02 PM


by: minnzen

Cool project. The React reconciler underneath Claude Code&#x27;s terminal layer is a solid foundation for this kind of real-time rendering.

4/1/2026, 6:06:16 PM


by: volume_tech

the blocking hooks observation matches what I would expect -- anything synchronous in the critical path has multiplicative effect when agents run 20-30 tool calls per task. even a 100ms write per call adds 2-3 seconds to a task, and that compounds across parallel agents fast.<p>th

4/1/2026, 6:07:05 PM


by: ivaivanova

Good to know background hooks make that much of a difference. How are you handling the case where multiple agent teams are writing to the same jsonl files simultaneously?

4/1/2026, 5:22:52 PM


by: theagentwall

great idea. I am curious what the future of coding with multiple terminals and agents will look like and this looks like a great start!

4/1/2026, 4:27:24 PM


by: nikita-ag

[dead]

4/1/2026, 6:18:13 PM


by: toolpipe_dev

[dead]

4/1/2026, 6:09:38 PM


by: jdurban

[dead]

4/1/2026, 5:28:20 PM


by: andrewmcwatters

[dead]

4/1/2026, 5:05:42 PM