微信扫一扫把 Skill 的源码、资源快照、README、包体和安装信号放进一个可搜索、可筛选的公开目录。
Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 fo…
Scalable data processing for ML workloads. Streaming execution across CPU/GPU, supports Parquet/CSV/JSON/images. Integrates with Ray Train, PyTorch, TensorFlow. Scales from single machine to 100s of n…
End-to-end paper reproduction from arxiv URL through smoke runs to replication experiments. Handles missing or partial official code, missing training scripts, missing hyperparameters, and private dat…
Index entry for an upstream research Skill intended for research work involving meta apply. Maintained upstream by wanshuiyin; source files are not included in this package.
Index entry for an upstream research Skill intended for research analysis related to the named method or subject. Maintained upstream by mims-harvard; source files are not included in this package.
>-
Audit experiment specs and empirical pipelines for claim-threatening validity and reproducibility failures.
Generate forest plots for meta-analysis. Renders publication-quality forest plots from synthesis data. Use when the user types `revman forest`.
Generate a Reveal.js HTML presentation from Markdown. Use when the user asks to generate a Reveal.js HTML presentation from Markdown.
Distributed training orchestration across clusters. Scales PyTorch/TensorFlow/HuggingFace from laptop to 1000s of nodes. Built-in hyperparameter tuning with Ray Tune, fault tolerance, elastic scaling.…
Index entry for an upstream research Skill intended for research analysis related to the named method or subject. Maintained upstream by ptreezh; source files are not included in this package.
Index entry for an upstream research Skill intended for research work involving meta optimize. Maintained upstream by wanshuiyin; source files are not included in this package.
Improve academic paper writing quality for ML/CV/NLP-style papers with clear section structure, paragraph flow, and reviewer-facing presentation. Use when drafting or revising Abstract, Introduction, …
Drafting point-by-point responses to peer reviewer comments for medical/biomedical journals. Covers tone calibration (firm but diplomatic), structuring the response letter, handling unreasonable comme…
RevMan data calculator for systematic reviews. Computes SD from SE/CI/IQR/range, calculates contrast data (OR/RR/RD/MD/SMD) from arm data, and combines study arms. Use when the user types `revman calc…
Generate funnel plots for assessing publication bias. Renders funnel plots with pseudo-95% CI lines and Egger's test. Use when the user types `revman funnel`.
>
Arbitrate conflicting reports into a source-grounded claim audit and reconciled synthesis.
Index entry for an upstream research Skill intended for research work involving feishu notify. Maintained upstream by wanshuiyin; source files are not included in this package.
Index entry for an upstream research Skill intended for research work involving stata accounting research. Maintained upstream by franklee16; source files are not included in this package.
Summarize deep research results into markdown report, cover all fields, skip uncertain values.
Use before marking any workflow stage as completed — verify outputs exist, are valid, and meet success criteria defined in workflow-state.json
Export systematic review as manuscript, data package, PRISMA checklist, or archive. Assembles all review data into publication-ready outputs. Use when the user types `revman export`.
Generate PRISMA 2020 flow diagrams for systematic reviews. Renders the study selection flowchart from screening numbers. Use when the user types `revman prisma`.