OS-Harm: A Benchmark for Measuring Safety of Computer Use Agents [NeurIPS 2025 Spotlight]
active 2025-06-19 → 2026-04-19 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 305 days from 2025-06-19 to 2026-04-19. Pushes: 3 total, peak 1 in a day. Pull requests: 3 total, peak 3 in a day. Issues: 2 total, peak 1 in a day. Comments: 5 total, peak 2 in a day. Stars: 32 total, peak 3 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| kevinhenryzz | 5 | 0 | 3 | 1 |
| max-andr | 4 | 2 | 0 | 2 |
| erfanshayegani | 3 | 0 | 0 | 2 |
| NielsRogge | 1 | 0 | 0 | 0 |
| thomas-kuntz | 1 | 1 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Pull request#8kevinhenryzz2026-01-17 19:19
- Issue comment#6kevinhenryzz2026-01-17 18:20Debug local judging script errors
- Pull request#6kevinhenryzz2026-01-17 18:19
- Pull request#5kevinhenryzz2026-01-17 17:32
- Issue comment#2erfanshayegani2025-07-09 22:05Severe slowdown after few runs
- Issue comment#2max-andr2025-07-09 08:37Severe slowdown after few runs
- Issue comment#2erfanshayegani2025-07-08 17:31Severe slowdown after few runs
- Issue#2erfanshayegani2025-07-07 22:17Severe slowdown after few runs
- Issue comment#1max-andr2025-06-19 09:41Release OS-Harm benchmark dataset on Hugging Face
- Issue#1NielsRogge2025-06-19 09:21Release OS-Harm benchmark dataset on Hugging Face
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 32 stars here means stars gained during the window, not the repo's star count.