Issues
Negative
Ratings
Loading…
POWER DRAIN< 4 VOLATILE4–5.4 FIELD-READY5.5–6.9 APEX7–8.5 OMEGA8.6–10 LIVE · cron every 20 min
- 8.6 – 10 OMEGA — battle-tested · deploy with confidence
- 7.0 – 8.5 APEX — production-ready · low risk
- 5.5 – 6.9 FIELD-READY — usable · keep an eye on issues
- 4.0 – 5.4 VOLATILE — known issues · deploy with care
- < 4.0 POWER DRAIN — significant problems · avoid for production
- 5.0 FLUX FIELD — neutral hold · brand-new release, signal still accumulating (first 3 hours)
Agent Watch tracks every OpenClaw and Hermes Agent release and scores it 0–10 for stability — built from real GitHub issues, LLM sentiment classification, and community ratings. Use the score to choose the safest version to deploy.
Inputs. Each release pulls in its referenced GitHub issues. An LLM classifies every issue as positive / negative / neutral with a confidence score. Each negative signal is weighted by recency, comment activity, and confidence. User ratings (1–10) blend in once submitted.
Formula. score = 10 · exp(−5 · bug_rate) where bug_rate = effective_negative_weight / max(24h, age_in_hours). Brand-new releases sit at neutral 5 for the first 3 hours while signal accumulates. Once the next 3 versions ship, the score is frozen against the latest input change so unchanged data doesn't drift.
Updated every 20 minutes via cron — pulled from public GitHub data + LLM classification.
syncing capability matrix