# I built an open-source AI coworker that logs in with 2FA without the model ever seeing your passwords

> Source: <https://dev.to/danielehrhardt/i-built-an-open-source-ai-coworker-that-logs-in-with-2fa-without-the-model-ever-seeing-your-48ep>
> Published: 2026-09-27 22:24:34+00:00

Every AI agent I tried had the same weak spot: it was great until it reached a login screen. Then it needed a password, a 2FA code, or me.

So I built **[Godmode Bot](https://github.com/codextde/godmode-bot)**, an open-source (MIT) AI coworker that has a real browser plus your logins and 2FA codes. The model never sees any of those secrets.

Godmode Bot is a desktop app for macOS, Windows and Linux built with Tauri 2. It can also run headless on a server, NAS or Raspberry Pi with a web dashboard. It uses [Claude Code](https://code.claude.com) as the brain and [browser-use](https://github.com/browser-use/browser-use) to drive a managed Chromium.

You can start with a single chat ("log into our billing portal and download September's invoice"), or build a team of persistent agents that each have their own instructions, memory, schedules and history.

This is the part I spent the most time on.

`vault_fill_login` or `vault_fill_totp`. Godmode types the value into the page over CDP. Getting secrets in is easy too. You can import passwords from Chrome, 1Password, Bitwarden, Apple Passwords or Firefox, and 2FA from a screenshot of a Google Authenticator *export* QR code (multi-account codes work).

Every agent gets its own repository:

```
~/.godmode/agents/invoice-collector/
├── CLAUDE.md                 # identity & instructions
├── MEMORY.md                 # long-term memory the agent maintains
├── conversations/<id>.md     # transcripts
├── runs/2026-09-27/<id>.jsonl# raw event logs (secrets redacted)
└── workspace/                # files the agent produced
```

Every run is committed, so you can see exactly what an agent learned and did, and roll it back.

Each turn runs `claude -p --output-format stream-json` inside the agent's git repo and streams every thought, tool call and screenshot to the UI. A local MCP gateway gives the agent vault tools, delegation tools and `report_missing_login`. You can watch the browser live next to the chat and take control at any point, for example to solve a CAPTCHA.

Agents run Claude Code with bypass permissions, so there are no permission prompts. Treat it like a trusted coworker with access to your machine. For sensitive setups, run it in a VM or container. [SECURITY.md](https://github.com/codextde/godmode-bot/blob/main/SECURITY.md) covers the threat model.

`docker compose up -d` and open I'd love feedback, especially on the vault design and what you'd want agents to do next. A full computer-use VM mode is on the roadmap.