evaluation
Tested projects carrying the evaluation topic, with their current verdicts. Topics come from each project's own metadata and Argusic's pool tags. Spellings of the same topic are gathered here under one name, so this page holds every project tagged any of them.
Measured by Argusic on a fresh machine every time. Every number links to its evidence.
What Argusic measured in evaluation
Argusic installed and ran 9 evaluation projects on clean machines: 7 reached a running state, 1 only ran against stand-in services, 1 did not run. Last tested November 16, 2026.
Every tested evaluation project
| subject | verdict | language | runs | last tested |
|---|---|---|---|---|
| Tracely-ai | Runs | Python | 2 | |
| ClawBench | Runs with mocks | Python | 2 | |
| trpc-agent-go | Runs | Go | 2 | |
| fable-method | Runs | Python | 2 | |
| yao-meta-skill | Runs | Python | 2 | |
| Hypha | Runs | TypeScript | 1 | |
| opik-openclaw | Runs | TypeScript | 1 | |
| langsmith-sdk | Runs | Python | 1 | |
| coze-loop | Could not verify | Go | 2 |
All tested projects: subjects.