speculative-historical-estimation
Provides confident, approximate answers to historical questions without standard disclaimers, using lists for multiple points, metric units, and specific numerical estimates rather than vague ranges.
historical-battle-analysis-with-canada-focus
Analyze historical battles by addressing specific criteria regarding participants, location, Canada's role, outcome, and significance. The output must be formatted in point form.
social-group-description-and-lingua-cultural-analysis
Analyzes a specific social group using a fixed 8-point demographic schema and V.I. Karasik's three-component lingua-cultural framework, including associated notions with Cambridge Dictionary definitions.
search-request-analysis-for-demographics
Act as an analyst to generate a list of top search requests for a specific demographic and topic based on search engine and social media analysis.
s3-to-excel-image-embedder-via-dataframe
Generates Python code to fetch images from AWS S3 using keys from a pandas DataFrame column and embeds them as visual images into an Excel file.
python-trading-signal-generator-no-ta-lib
Generates a Python function `signal_generator` that uses pandas for EMA/SMA and candlestick analysis to produce Buy/Sell signals, excluding TA-Lib and trend calculation.
eda-iterative-place-and-route-flow
Generates Python code for an EDA workflow that iteratively places and routes a netlist, checking design rules and stopping when quality targets are met.
kohlcpcapca
针对金融K线OHLC数据进行降维分析，包括数据预处理（变化率计算、标准化）、标准PCA降维以及基于随机化SVD的伪PCA降维的Python实现步骤。
python-pandas
根据指定的行切片规则和列映射，将源DataFrame（df_sub）的列数据转换为长格式DataFrame，包含sub_id、trial、start_time、end_time等字段。
flaskmatplotlib
使用Flask、Matplotlib、APScheduler和Flask-Caching构建实时数据可视化应用。图像在后台定期更新并存储在内存缓存中，不保存到本地磁盘，通过HTTP接口提供给前端。
dataframemysql
用于将Pandas DataFrame同步到MySQL数据库的技能，包含针对JSON字段、数值类型（int/float）、字符串顺序（如effect_desc）的归一化比较逻辑，以及基于type字段的条件排除规则。
standard-release-process
General SOP for common requests related to diffusion, 2023, stable.
tf-agents-lstm-multi-stock-training
配置TF-Agents的DQN代理使用自定义LSTM网络处理多只股票的时间序列数据，涵盖环境批量打包、维度适配、网络初始化避坑以及完整的训练与评估循环，兼容TensorFlow 2.10.1。
tensorflow-java-savedmodel
使用 TensorFlow Java API 0.4.0 加载 SavedModel 格式的模型，处理三维输入数据并进行预测。包含数据类型转换、Tensor 初始化、资源管理及输入输出节点名称匹配的完整流程。
calculate-aligned-partition-boundaries
Calculates optimal partition start and end points based on 1MB alignment constraints to maximize space, or provides concise parted commands for alignment.
clean-database-output-for-excel
Transforms raw database query results into professional, Excel-ready tables by removing metadata and rounding numbers.
generate-aspnet-mvc-entity-framework-models-from-schema
Generates C# model classes with specific data annotations based on a provided database schema, ensuring Primary Keys, Unique constraints, Foreign Keys, and specific data types (like varchar or byte arrays) are correctly implemented.
generate-spring-boot-jdbc-multiple-update-stack
Generates the Controller, Service, and DAO layers for a multiple records update operation using Spring Boot and JdbcTemplate, transforming a provided delete pattern into an update pattern.
dax-date-intersection-count-with-rankx
Create a DAX calculated measure to count the number of rows where the StartDate is less than the EndDate of the previous row, partitioned by Customer and SKU and ordered by StartDate using RANKX.
excel-vba-double-click-task-duplication-macro
Generates a VBA Worksheet_BeforeDoubleClick event handler to duplicate data rows, shift content down, clear the immediate next row, and prevent cell edit mode based on specific user requirements.
item-based-collaborative-filtering-movie-recommender
Build a Python model to recommend the top 10 similar movies using item-based collaborative filtering for a dataset with a specific 3-column schema (movie_id, title with year, pipe-separated genres).