FINDING #6248 · 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)
10% OF STARS IN ARCHIVE
Growth Telemetry
VELOCITY /D
5.73
ACCEL
+0.03
RETENTION
30.6%
PEAK 2026-10-09 · FORK-RETENTION 17.6% · 172 STARS / WINDOW
Author Audience
AUDIENCE
3,101
FOLLOWERS
385
OWNER ★
11,657
Engagement Signals
FORKS
189
ISSUE AUTH
0
PR AUTH
0
UNIQUE STARGAZERS 172 / 172 (DIVERSITY 1.00)
Why This Is A Finding
AMAP-ML/LongHorizon-Harness собрал 172 звёзд за окно, тогда как у автора всего 385 подписчиков — эффективная аудитория ≈ 3,101. Это даёт surprise-индекс 0.00183 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 17.6% и 0 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация положительная — рост ещё не выдохся.
Related Findings
RANKS ABOVE 35% OF 9675 FINDINGS
METRICS IN CONTEXT
MEDIAN ACROSS ALL 9675 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW
METRICVALUEMEDIANΔ MEDPERCENTILE
SCORE0.000.00-0.00ABOVE 35%
VELOCITY5.733.80+1.93ABOVE 64%
RETENTION30.6%20.3%+10.3 PPABOVE 72%
FORKS189138+51ABOVE 57%
SURPRISE0.000.01-0.01ABOVE 22%