FINDING #4642 · 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)
1% OF STARS IN ARCHIVE
Growth Telemetry
VELOCITY /D
7.67
ACCEL
+2.50
RETENTION
0.0%
PEAK 2026-10-12 · FORK-RETENTION 0.0% · 23 STARS / WINDOW
Author Audience
AUDIENCE
3,070
FOLLOWERS
383
OWNER ★
11,518
Engagement Signals
FORKS
188
ISSUE AUTH
0
PR AUTH
0
UNIQUE STARGAZERS 23 / 23 (DIVERSITY 1.00)
Why This Is A Finding
AMAP-ML/LongHorizon-Harness собрал 23 звёзд за окно, тогда как у автора всего 383 подписчиков — эффективная аудитория ≈ 3,070. Это даёт surprise-индекс 0.00247 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 0.0% и 0 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация положительная — рост ещё не выдохся.
Related Findings
RANKS ABOVE 42% OF 8038 FINDINGS
METRICS IN CONTEXT
MEDIAN ACROSS ALL 8038 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW
METRICVALUEMEDIANΔ MEDPERCENTILE
SCORE0.000.00-0.00ABOVE 42%
VELOCITY7.675.67+2.00ABOVE 60%
RETENTION0.0%14.3%-14.3 PPABOVE 0%
FORKS188333-144ABOVE 39%
SURPRISE0.000.01-0.00ABOVE 38%