My Agents Do the Work, I Just Press "Allow" 🦥

Description and use

AgentWebUI consolidates approval interfaces for multiple AI coding agents into a single web UI with an Allow button, simplifying management and reducing token usage.

Problem

Multiple AI coding agents require separate approval interfaces, making it hard to manage approvals consistently.

Outcome

AgentWebUI consolidates approval workflows for multiple coding agents into a single, consistent interface, reducing overhead and token costs by leveraging local models for simple queries.

Workflow

• Install AgentWebUI via git clone and docker compose
• Configure .env with desired models and agents
• Open the web UI in a browser
• Select an agent, model, effort level, and target folder
• Watch the agent’s plan and diffs in real time
• Press Allow or Deny when the agent requests file changes

Evidence

At one point last month I had four coding agents installed, and each of them asked for permission in its own way.

Source

So the three of us at HackersHostel built one place where that button always shows up in the same spot.

Source

AgentWebUI is a small, self-hosted web page for talking to AI models and command-line coding agents from one place.

Source

With an agent, you choose the agent, then its model and effort level, then the folder it should work in. After that you mostly watch. It streams the agent's thinking, its plan, the tools it calls and the diffs it makes. When it wants to touch a file, it asks first. You read the diff and press Allow or Deny.

Source

There are no accounts and no setup wizard. Everything lives in one .env file.

Source

The server is about 3,000 lines of Python with one dependency (aiohttp) and uses about 40 MB of memory.

Source

The page itself loads nothing from the internet unless you pick a font other than Geist.

Source

I don't open a terminal just to check whether an agent is stuck waiting for me. The request shows up in the browser, with the diff right there.

Source