Light 科研主线 stage 6:把冻结的 question/estimand、experiment matrix、pre-registration 与 data lineage 落成最小可运行、测试先行、无泄漏、可复现且能交给 re…
Part of Light SkillsPack · 23 skills
by Light0305· ResearchUnclaimed · Yours? Claim it
About
Light 科研主线 stage 6:把冻结的 question/estimand、experiment matrix、pre-registration 与 data lineage 落成最小可运行、测试先行、无泄漏、可复现且能交给 result-analysis 的实验代码。用于实现或复现训练/预处理/评测, 设计 gold/property/metamorphic 测试,控制 Python/NumPy/框架/CUDA/DataLoader 随机性,审查 train/test 或 CV fit 穿越,记录 config/code/environment/input hashes、stdout/stderr、raw metrics、patient/entity predictions 与 failure artifacts,以及运行 stage-6 checkpoint。数据泄漏或不可复现是 critical;静态扫描和同 seed 两次一致都不证明跨硬件 绝对复现。
Skills people install alongside light-experiment-coding.
Runs on your machine.
Runs on your machine.
Runs on your machine.
Runs on your machine.
npx -y degit Light0305/Light-skills/skills/light-experiment-coding ~/.copilot/skills/light-experiment-coding
Runs on your machine.
npx -y degit Light0305/Light-skills/skills/light-experiment-coding ~/.cursor/skills/light-experiment-coding
Runs on your machine.
npx -y degit Light0305/Light-skills/skills/light-experiment-coding ~/.claude/skills/light-experiment-coding
Runs on your machine.
npx -y degit Light0305/Light-skills/skills/light-experiment-coding ~/.agents/skills/light-experiment-coding
Runs on your machine.
Runs on your machine.
npx -y degit Light0305/Light-skills/skills/light-experiment-coding ~/.gemini/skills/light-experiment-coding
Runs on your machine.
npx -y degit Light0305/Light-skills/skills/light-experiment-coding .agents/skills/light-experiment-coding
Runs on your machine.
npx -y degit Light0305/Light-skills/skills/light-experiment-coding ~/.openclaw/skills/light-experiment-coding
Runs on your machine.
npx -y degit Light0305/Light-skills/skills/light-experiment-coding ~/.config/opencode/skills/light-experiment-coding
Runs on your machine.
npx -y degit Light0305/Light-skills/skills/light-experiment-coding ~/.hermes/skills/light-experiment-coding
Scanned and passed · checked 2 h ago
From people who use it
Step by step
Nothing new reaches your agent until a person has looked at it.
Scanned on every change, and again when our rules improve.
No reviews yet. Used light-experiment-coding? Say how it went.
Write a reviewSkills people install alongside light-experiment-coding.
Khazix's everyday skills: goal briefs for long agent runs, disk cleanup, AI news, doc tidy-ups, deep research and Chinese long-form writing.
A library of AI research skills that lets your agent fine-tune, train, evaluate and serve models, and take a project from idea to paper.
A six-step competitive analysis for product managers, from the market landscape to rivals' strategic direction, ending in battle cards.
Researches tech and advanced-manufacturing investment themes by finding supply-chain bottlenecks, then ranks which companies to study first.
Nobody has shared how they use light-experiment-coding yet.
Security scan
Passed
Safe to install in every agent it supports. Runs in your agent, as instructions.