数据分析、机器学习、LLM 应用、提示词设计（9088 个）
Use this skill to understand image contents with GPT-4 Vision. It can describe a picture, extract text from screenshots, identify visible objects, and answer questions about an image, returning the result in a structured format.
中文简介正在准备中。当前显示英语简介。
Support visual inspection and management in industrial settings across more than 20 sectors, including retail, energy, mining, ports, chemicals, and steel. The provided script can analyze an image or generate a report using Baidu Qianfan AppBuilder.
中文简介正在准备中。当前显示英语简介。
Manage algorithm experiments, adjust parameters, and review visualized results while working to optimize complex algorithms. It is intended for tasks that involve iterative algorithm improvement and experiment tracking, and its included script requires an AppBuilder token.
中文简介正在准备中。当前显示英语简介。
Use Apify Actors, prebuilt web scraping tools, to collect structured data from social media, search engines, ecommerce sites, and other pages. The skill selects an Actor for the request and can return results as JSON or CSV, with rate limits and ethical scraping guidance.
中文简介正在准备中。当前显示英语简介。
Plan a parsing workflow for repeated text such as forms or quizzes. Use regular expressions for predictable patterns, score extraction confidence, and reserve language model checks for uncertain edge cases to help control cost.
中文简介正在准备中。当前显示英语简介。
Use this skill to design ClickHouse tables, improve analytical queries, and shape data ingestion and ETL workflows. It provides practical patterns for aggregations, materialized views, monitoring, and common analyses such as time series, retention, and funnels.
中文简介正在准备中。当前显示英语简介。
Design LLM API pipelines that manage spending through task-based model selection and budget tracking. The skill also shows how to retry transient failures with backoff and cache long system prompts, while keeping cost records auditable.
中文简介正在准备中。当前显示英语简介。
Design ClickHouse tables and write faster analytical queries for large datasets. The skill covers MergeTree choices, partitioning and sort keys, batch ingestion, materialized views, performance checks, and common analyses such as time series, funnels, and cohorts.
中文简介正在准备中。当前显示英语简介。
Use this skill to design ClickHouse tables, improve analytical queries, and build data pipelines. It provides patterns for bulk and streaming inserts, materialized views, performance monitoring, and common analyses such as retention and funnels.
中文简介正在准备中。当前显示英语简介。
Use this pattern to help sub-agents find the code and project context they need. It starts with a broad search, rates file relevance, identifies gaps, and refines the query for up to three cycles.
中文简介正在准备中。当前显示英语简介。
This skill describes a Claude Code learning system that observes sessions through hooks and records small, confidence-rated behavior patterns called instincts. It keeps project-specific patterns separate from global ones, and can group instincts into reusable skills, commands, or agents.
中文简介正在准备中。当前显示英语简介。
Plan a parser for repeated, structured text such as forms or quizzes. Start with regular expressions, score extraction confidence, and reserve large language model checks for uncertain cases that need review.
中文简介正在准备中。当前显示英语简介。
Apply ClickHouse patterns to analytical data work, from MergeTree table design and query optimization to bulk or streaming ingestion, materialized views, and performance checks. It also covers time-series, funnel, retention, and cohort analyses for large datasets.
中文简介正在准备中。当前显示英语简介。
Use this skill to design LLM API workflows that track spending, choose models based on task complexity, retry temporary failures, and cache repeated prompts. It outlines how to combine these practices into a reusable pipeline with budget checks.
中文简介正在准备中。当前显示英语简介。
A low-level plotting library for comprehensive customization. Use when fine-grained control over every plot element is needed, creating new types of charts, or integrating into specific scientific workflows. Can export to PNG/PDF/SVG for publication. For quick statistical charts, use seaborn; for interactive charts, use plotly; for journal-style, publication-ready multi-panel charts, use scientific-visualization.
中文简介正在准备中。当前显示原文说明。
Prepare research analysis code for publication or collaboration by documenting its functions, replacing hardcoded settings with parameters, adding error handling, and specifying dependencies. The workflow also checks whether refactoring preserves behavior and supports repeatable results.
中文简介正在准备中。当前显示英语简介。
Generate Circos configuration files for circular genomics data visualization. Supports genomic variations (SNPs, CNVs, structural variants), cell-cell communication networks, and custom track configurations for publication-ready circular plots.
中文简介正在准备中。当前显示原文说明。
Clean and prepare tabular data with pandas and numpy. Handle missing values and duplicates, reshape and filter tables, merge datasets, scale numeric columns, convert types, aggregate results, and create derived features for analysis or later modeling.
中文简介正在准备中。当前显示英语简介。
Create a forest plot for a meta-analysis of continuous outcomes from a CSV of study means, standard deviations, and group sizes. The skill checks the input and produces a PNG chart, a CSV data table, and a summary of pooled effects and heterogeneity.
中文简介正在准备中。当前显示英语简介。
Use lab budget forecaster for data analysis workflows that need structured execution, explicit assumptions, and clear output boundaries.
中文简介正在准备中。当前显示原文说明。
Statistical analysis and reporting for experimental datasets; use when you need to interpret experimental results, test significance (t-tests/ANOVA), or generate reproducible reports.
中文简介正在准备中。当前显示原文说明。
Use facs gating viz style for data analysis workflows that need structured execution, explicit assumptions, and clear output boundaries.
中文简介正在准备中。当前显示原文说明。
Explore scientific data files across more than 200 formats. The skill identifies file types, summarizes structure and statistics, checks data quality, and recommends visualizations or next analyses in a Markdown report. Specialized libraries may be needed for some formats.
中文简介正在准备中。当前显示英语简介。
Perform quantitative meta-analysis with effect size calculation, forest plots, funnel plots, and heterogeneity assessment. Use when: user asks to combine results from multiple studies, calculate pooled effect sizes, assess publication bias, or create forest/funnel plots. NOT f...
中文简介正在准备中。当前显示原文说明。