FINDING #8028 · UNIT ID 1335115548
dltsum/research-guard
Fail-closed cross-disciplinary research design, novelty, evidence, writing, and audit plugin for Codex-compatible agents.
SURPRISE SCORE
0.00
Score Breakdown
SCORE = SURPRISE × ENGAGEMENT^0.7 × FRESHNESS × VISIBILITY × CONFIDENCE
SURPRISE = WINDOW STARS / DAYS / (AUDIENCE + FLOOR)
57% OF STARS IN ARCHIVE
[BOT] SUSPECTED STAR BOT — SCORE PENALIZED. SIGNATURES:
S2 · NO EXTERNAL ISSUE/PR AUTHORS DESPITE 100+ STARS
S5 · PREDATES WINDOW, YET HALF+ OF ALL ITS STARS LANDED IN IT
Growth Telemetry
VELOCITY /D
13.14
ACCEL
-0.54
RETENTION
53.6%
PEAK 2026-09-08 · FORK-RETENTION 0.0% · 92 STARS / WINDOW
Author Audience
AUDIENCE
31
FOLLOWERS
7
OWNER ★
240
Engagement Signals
FORKS
1
ISSUE AUTH
0
PR AUTH
0
UNIQUE STARGAZERS 92 / 92 (DIVERSITY 1.00)
Why This Is A Finding
dltsum/research-guard собрал 92 звёзд за окно, тогда как у автора всего 7 подписчиков — эффективная аудитория ≈ 31. Это даёт surprise-индекс 0.19 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 0.0% и 0 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация отрицательная — внимание остывает после пика.
Related Findings
RANKS ABOVE 18% OF 9789 FINDINGS
METRICS IN CONTEXT
MEDIAN ACROSS ALL 9789 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW
METRICVALUEMEDIANΔ MEDPERCENTILE
SCORE0.000.00-0.00ABOVE 18%
VELOCITY13.143.43+9.71ABOVE 84%
RETENTION53.6%37.5%+16.1 PPABOVE 74%
FORKS1120-119ABOVE 2%
SURPRISE0.190.01+0.18ABOVE 98%