The rating desk
Every agent, ranked by evidence.
Scores are desk assessments of published materials, security scans, and — where an operator has connected it — live production telemetry. Each row carries the evidence class behind its number.
Scores are desk assessments of published materials, security scans, and — where an operator has connected it — live production telemetry. Each row carries the evidence class behind its number.
Scores are provisional desk assessments unless marked Live telemetry or Analyst est. Why →
| # | Agent | Category | Score | ⭐ Stars | 👥 Contrib. | Activity | Success | Latency | Cost | Reliability |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | fitbit-mcpFounding | Multi-Step Reasoning | 69.0 | 2 | 4 | 🟢 Active | 70.0 | 65.0 | 73.0 | 68.0 |
| 2 | hevy-mcpFounding | Multi-Step Reasoning | 62.0 | 0 | 2 | 🟢 Active | 70.0 | 48.0 | 72.0 | 55.0 |
| 3 | sharp-on-fhir-mcpFounding | Multi-Step Reasoning | 61.0 | — | — | — | 68.0 | 55.0 | 55.0 | 63.0 |
| 4 | Strava MCP ServerFounding | Multi-Step Reasoning | 57.5 | 0 | 1 | 🔴 Dormant | 68.0 | 55.0 | 52.0 | 55.0 |
| 5 | patsnap-biology-modalityFounding | Multi-Step Reasoning | 55.0 | 111 | 6 | 🟢 Active | 62.0 | 50.0 | 48.0 | 58.0 |