[API] Stand up Computer Use as a browser automation harness #755

Closed
opened 2026-03-30 01:56:57 +00:00 by Timmy · 1 comment
Owner

Phase 5 — Advanced Integration

Objective

Use the Gemini Computer Use API to let Timmy perform browser-based tasks autonomously (web research, form filling, screenshots, etc.).

Use Cases

  • Automated testing of The Nexus in a real browser
  • Web research tasks that require interaction (not just reading)
  • Automated screenshot capture of deployed Nexus for visual regression
  • Filling out forms on external services on Timmy's behalf

Acceptance

  • Set up a sandboxed browser environment for Computer Use
  • Implement a BrowserAgent class in sovereign-orchestration
  • Define safety boundaries: allowlist of sites, no financial transactions
  • Log all browser actions with screenshots for audit
  • Test: Timmy can autonomously navigate to The Nexus and report status

Safety

This is a powerful capability. Must have:

  • Explicit operator approval for new sites
  • Action logging with screenshots
  • Kill switch accessible via Hermes

Refs

  • Computer Use API docs (Google)
## Phase 5 — Advanced Integration ### Objective Use the Gemini Computer Use API to let Timmy perform browser-based tasks autonomously (web research, form filling, screenshots, etc.). ### Use Cases - Automated testing of The Nexus in a real browser - Web research tasks that require interaction (not just reading) - Automated screenshot capture of deployed Nexus for visual regression - Filling out forms on external services on Timmy's behalf ### Acceptance - [ ] Set up a sandboxed browser environment for Computer Use - [ ] Implement a `BrowserAgent` class in sovereign-orchestration - [ ] Define safety boundaries: allowlist of sites, no financial transactions - [ ] Log all browser actions with screenshots for audit - [ ] Test: Timmy can autonomously navigate to The Nexus and report status ### Safety This is a powerful capability. Must have: - Explicit operator approval for new sites - Action logging with screenshots - Kill switch accessible via Hermes ### Refs - Computer Use API docs (Google)
Timmy added this to the M5: Google AI Ultra Integration milestone 2026-03-30 01:56:57 +00:00
Timmy added the google-ai-ultraharnessp2-backloggemini-api labels 2026-03-30 01:56:57 +00:00
Timmy changed title from [API] Integrate Computer Use API for browser-based agent automation to [API] Stand up Computer Use as a browser automation harness 2026-03-30 02:56:22 +00:00
Author
Owner

Audit: Google AI Ultra integration epic — these are aspirational proposals, not scoped work. Closing. Reopen individually with acceptance criteria if needed.

Audit: Google AI Ultra integration epic — these are aspirational proposals, not scoped work. Closing. Reopen individually with acceptance criteria if needed.
Timmy closed this issue 2026-04-03 22:59:57 +00:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Timmy_Foundation/the-nexus#755