Agent Browser vs Cua
An independent, evidence-based trust comparison of Agent Browser and Cua, two Browser & Computer Use projects in the HVTracker registry. Scores come from public, checkable signals — supply-chain provenance, OSSF Scorecard, maintenance, and adoption — not popularity.
Agent Browser leads on trust — 86.7/100 (Grade A) vs 68.8/100 (Grade B), a 17.9-point gap. Full breakdown below.
| Signal | Agent Browservercel-labs/agent-browser | Cuatrycua/cua |
|---|---|---|
| HVTrust score | 86.7 | 68.8 |
| Evidence grade | A | B |
| Coverage grade | B | B |
| Overall rank | #26 | #119 |
| Rank in Browser & Computer Use | #2 | #7 |
| GitHub stars | 39.9k | 20.9k |
| Last updated | today | today |
| Build provenance | Yes | No |
| OSSF Scorecard | 4.3 / 10 | 3.0 / 10 |
| License | Apache-2.0 | MIT |
| Downloads | 923k/wk | 11k/wk |
| Trust dimensions (points earned) | ||
| Safety / integrity / 25 | 17.9 | 7.0 |
| Identity & provenance / 18 | 18.0 | 10.8 |
| Transparency / 17 | 12.2 | 11.1 |
| Maintenance / 20 | 18.0 | 20.0 |
| Adoption / 20 | 19.0 | 15.8 |
| Runtime capability surface (full matrix) | ||
| MCP server | Implemented | Implemented |
| External providers | 3 — Amazon Bedrock, Browserbase, Redis | 3 — Amazon Bedrock, Anthropic, OpenAI |
| Requires API keys | Yes | No |
| Plugin surface | extensions | — |
| Provenance drift | Match | Match |
How to read this: HVTrust (0–100) weighs supply-chain signals (provenance, OSSF Scorecard, signed commits, open license) alongside real-world adoption, scaled by an evidence-confidence factor. Grade bands: A ≥ 80, B ≥ 65, C ≥ 50, D < 50. Signals refresh daily. Full methodology v4.2 →