Loading release run…
Loading release run…
NousResearch/hermes-agent · v2026.6.5…v2026.6.19
LLM-as-judge rubric: average 3.75 / 5 across 4 artifacts.
| Metric | Score | Detail |
|---|---|---|
| Evidence coverage | n/a | 0 / 0 |
| Unsupported-claim rate | n/a | 0 / 0 |
| Edit distance (reviewer rewrite) | n/a | 0 samples |
| Approval latency | n/a | 0 samples |
| Notify-to-decision latency | n/a | 0 samples |
| Feature rejection rate | 0.0% | 0 / 3 |
| Skill-candidate acceptance rate | 100.0% | 1 / 1 (repo-wide) |
| Media success rate | 100.0% | 2 / 2 |
| Engagement: views (run total) | not yet reported | — |
| Engagement: clicks (run total) | not yet reported | — |
| Engagement: conversions (run total) | not yet reported | — |
LLM-as-judge average per dimension, 1 (poor) to 5 (excellent), across this run’s approved artifacts.
| Dimension | Avg score |
|---|---|
| Claim support | 3.00 / 5 |
| Claim risk (5 = low) | 4.00 / 5 |
| Brand voice | 4.25 / 5 |
| Audience relevance | 4.00 / 5 |
| Originality | 3.00 / 5 |
| Conversion intent | 3.75 / 5 |
| Clarity | 5.00 / 5 |
| Demoability | 3.00 / 5 |