Fence vs Rivet Agents
Both are Grade C and 3.1 points apart, so choose on what you weigh most. Fence leads on supply-chain integrity and transparency; Rivet Agents leads on maintenance and adoption.
Lightweight, container-free sandbox for running commands with network and filesystem restrictions
Choose Fence if supply-chain integrity and transparency matter most.
- +6.6Safety / Integrity: 79% of recent commits signed, against 35%
- +2.9Transparency: OSSF Scorecard 7.0 against 3.5
A durable process per agent, with memory that survives restarts
Choose Rivet Agents if maintenance and adoption matter most.
- +6.8Maintenance: last push 2d ago, against 7d ago
- +0.5Adoption: 1.6k GitHub stars against 1k
Where they differ
1 dimension identical: Identity 10.8 · Full evidence table
An independent, evidence-based trust comparison of Fence and Rivet Agents, two Sandboxes & Runtimes projects in the HVTracker registry. Scores come from public, checkable signals — supply-chain provenance, OSSF Scorecard, maintenance, and adoption — not popularity.
Full evidence
| Signal | Fencefencesandbox/fence | Rivet Agentsrivet-dev/agents |
|---|---|---|
| HVTrust score | 56.6 | 53.5 |
| Evidence grade | C | C |
| Coverage grade | C | C |
| Overall rank | #801 | #900 |
| Rank in Sandboxes & Runtimes | #6 | #7 |
| GitHub stars | 1.0k | 1.6k |
| Last updated | 7d ago | 2d ago |
| Build provenance | No | No |
| OSSF Scorecard | 7.0 / 10 | 3.5 / 10 |
| License | Apache-2.0 | Apache-2.0 |
| Downloads | — | — |
| Trust dimensions (points earned) | ||
| Safety / integrity / 25 | 12.7 | 6.1 |
| Identity & provenance / 18 | 10.8 | 10.8 |
| Transparency / 17 | 14.4 | 11.5 |
| Maintenance / 20 | 11.5 | 18.3 |
| Adoption / 20 | 7.2 | 7.7 |
| Runtime capability surface (full matrix) | ||
| MCP server | — | Declared |
| External providers | — | 1 — E2B |
| Requires API keys | No | No |
| Plugin surface | — | extensions |
| Provenance drift | — | — |
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 v4.4 →