FINDING #1306 · UNIT ID 1322509558
AMAP-ML/LongHorizon-Harness
The long-horizon computer-use harness. Run AI agents across desktop apps and the CLI for extended periods while preserving task state and making reliable progress on complex workflows. Features fresh-context execution, durable verified state, independent auditing, recoverable progress, and native Claude Code / Codex / OpenClaw integration.
SURPRISE SCORE
0.00
Score Breakdown
SCORE = SURPRISE × ENGAGEMENT^0.7 × FRESHNESS × VISIBILITY × CONFIDENCE
SURPRISE = WINDOW STARS / DAYS / (AUDIENCE + FLOOR)
69% OF STARS IN ARCHIVE
Growth Telemetry
VELOCITY /D
55.86
ACCEL
+4.39
RETENTION
55.2%
PEAK 2026-08-05 · FORK-RETENTION 75.9% · 391 STARS / WINDOW
Author Audience
AUDIENCE
2,457
FOLLOWERS
317
OWNER ★
9,117
Engagement Signals
FORKS
67
ISSUE AUTH
0
PR AUTH
0
UNIQUE STARGAZERS 391 / 391 (DIVERSITY 1.00)
Why This Is A Finding
AMAP-ML/LongHorizon-Harness собрал 391 звёзд за окно, тогда как у автора всего 317 подписчиков — эффективная аудитория ≈ 2,457. Это даёт surprise-индекс 0.0224 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 75.9% и 0 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация положительная — рост ещё не выдохся.
Related Findings
RANKS ABOVE 87% OF 9721 FINDINGS
METRICS IN CONTEXT
MEDIAN ACROSS ALL 9721 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW
METRICVALUEMEDIANΔ MEDPERCENTILE
SCORE0.010.00+0.01ABOVE 87%
VELOCITY55.863.43+52.43ABOVE 96%
RETENTION55.2%32.1%+23.0 PPABOVE 84%
FORKS67130-63ABOVE 34%
SURPRISE0.020.01+0.02ABOVE 78%