An independent, evidence-based trust comparison of Strands Agents and Vercel AI SDK, two Agent Frameworks projects in the HVTracker registry. Scores come from public, checkable signals — supply-chain provenance, OSSF Scorecard, maintenance, and adoption — not popularity.
| Signal | Strands Agentsstrands-agents/harness-sdk | Vercel AI SDKvercel/ai |
|---|---|---|
| HVTrust score | 91.7 | 95.1 |
| Evidence grade | A | A |
| Coverage grade | B | A |
| Overall rank | #4 | #1 |
| Rank in Agent Frameworks | #3 | #1 |
| GitHub stars | 6.5k | 25.5k |
| Last updated | 1d ago | 1d ago |
| Build provenance | Yes | Yes |
| OSSF Scorecard | 5.6 / 10 | 7.3 / 10 |
| License | Apache-2.0 | NOASSERTION |
| Downloads | 7.8M/wk | 15.3M/wk |
| Trust dimensions (points earned) | ||
| Safety / integrity / 25 | 19.5 | 21.6 |
| Identity & provenance / 20 | 18.0 | 18.0 |
| Transparency / 17 | 13.3 | 14.7 |
| Maintenance / 20 | 19.9 | 19.9 |
| Adoption / 20 | 18.3 | 20.0 |
| Runtime capability surface (full matrix) | ||
| MCP server | Implemented | Implemented |
| External providers | — | 2 — Anthropic, OpenAI |
| Requires API keys | No | 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 →