Agent Infra Sandbox vs OpenShell
An independent, evidence-based trust comparison of Agent Infra Sandbox and OpenShell, two Sandboxes & Runtimes projects in the HVTracker registry. Scores come from public, checkable signals — supply-chain provenance, OSSF Scorecard, maintenance, and adoption — not popularity.
OpenShell leads on trust — 70.6/100 (Grade B) vs 54.6/100 (Grade C), a 16.0-point gap. Full breakdown below.
| Signal | Agent Infra Sandboxagent-infra/sandbox | OpenShellNVIDIA/OpenShell |
|---|---|---|
| HVTrust score | 54.6 | 70.6 |
| Evidence grade | C | B |
| Coverage grade | C | B |
| Overall rank | #793 | #320 |
| Rank in Sandboxes & Runtimes | #9 | #5 |
| GitHub stars | 5.9k | 8.7k |
| Last updated | 4d ago | today |
| Build provenance | No | No |
| OSSF Scorecard | 4.6 / 10 | 6.5 / 10 |
| License | Apache-2.0 | Apache-2.0 |
| Downloads | — | 25k/wk |
| Trust dimensions (points earned) | ||
| Safety / integrity / 25 | 9.9 | 13.1 |
| Identity & provenance / 18 | 10.8 | 10.8 |
| Transparency / 17 | 12.4 | 14.0 |
| Maintenance / 20 | 13.4 | 20.0 |
| Adoption / 20 | 9.1 | 15.3 |
| Runtime capability surface (full matrix) | ||
| MCP server | Declared | — |
| External providers | 1 — OpenAI | 3 — Anthropic, OpenAI, OpenRouter |
| Requires API keys | Yes | Yes |
| Plugin surface | — | extensions |
| Provenance drift | — | Unknown |
Open in the live compare tool →
Agent Infra Sandbox profile
OpenShell profile
More Sandboxes & Runtimes →
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 v4.3 →