wanshuiyin/Auto-claude-code-research-in-sleep — GitHub trend

FINDING #91 · UNIT ID 1177608268

wanshuiyin/Auto-claude-code-research-in-sleep

ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

[ PYTHON ][ GITHUB ↗ ]

SURPRISE SCORE

0.00

Score Breakdown

SURPRISE0.00542

ENGAGEMENT1.46

FRESHNESS1.54

SCORE = SURPRISE × ENGAGEMENT^0.7 × FRESHNESS × VISIBILITY × CONFIDENCE

SURPRISE = WINDOW STARS / DAYS / (AUDIENCE + FLOOR)

12% OF STARS IN ARCHIVE

Growth Telemetry

VELOCITY /D

9.19

ACCEL

-0.03

RETENTION

11.1%

PEAK 2026-03-14 · FORK-RETENTION 88.6% · 1,655 STARS / WINDOW

Author Audience

AUDIENCE

1,656

FOLLOWERS

263

OWNER ★

13,929

Engagement Signals

FORKS

1,212

ISSUE AUTH

PR AUTH

UNIQUE STARGAZERS 1,652 / 1,655 (DIVERSITY 1.00)

Why This Is A Finding

wanshuiyin/Auto-claude-code-research-in-sleep собрал 1,655 звёзд за окно, тогда как у автора всего 263 подписчиков — эффективная аудитория ≈ 1,656. Это даёт surprise-индекс 0.00542 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 88.6% и 62 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация отрицательная — внимание остывает после пика.

Related Findings

RANKS ABOVE 97% OF 2987 FINDINGS

#03originalankur/maptoposter[ PYTHON ]0.0234 #05HeartMuLa/heartlib[ PYTHON ]0.0222 #08sickn33/agentic-awesome-skills[ PYTHON ]0.0207 #14cft0808/edict[ PYTHON ]0.0183 #17aden-hive/hive[ PYTHON ]0.0175

METRICS IN CONTEXT

MEDIAN ACROSS ALL 2987 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW

METRICVALUEMEDIANΔ MEDPERCENTILE

SCORE0.010.00+0.01ABOVE 97%

VELOCITY9.192.84+6.35ABOVE 87%

RETENTION11.1%6.8%+4.3 PPABOVE 68%

FORKS1,2121,068+144ABOVE 53%

SURPRISE0.010.00+0.00ABOVE 91%