Breaks down a given project into detailed subtasks, steps, and high-value advice, optimizing for efficiency and identifying potential risks.
Skills(SKILL.md)は、AIエージェント(Claude Code、Cursor、Codexなど)に特定の能力を追加するための設定ファイルです。
詳しく見る →Breaks down a given project into detailed subtasks, steps, and high-value advice, optimizing for efficiency and identifying potential risks.
Provides concise answers and reasons for user-provided statements or questions, adhering to a short format.
Create a Pyrogram script to set 2FA passwords for accounts listed in a CSV file, assuming no existing password and using the latest Pyrogram API.
Generates a complete Python project featuring a Tkinter GUI and a Streamlit web dashboard for a bar stock exchange system. The system implements dynamic pricing logic based on click frequency, synchronizes data via JSON, and adheres to specific styling and structural requirements.
Generates a Python script to simulate a terminal session where status messages appear with random processing delays and commands are printed character-by-character to mimic human typing.
Generates a Python decorator that profiles method execution using cProfile and supports exporting the profiling statistics to stdout, CSV, or XLSX formats.
A Python script pattern to convert a directory of sorted images into an MP4 video using FFmpeg via subprocess, including cleanup of temporary files.
Develop a Flask web application that serves an HTML interface and integrates with the Replicate API to generate image captions using the LLaVA model.
Generates a complete Python Flask CRM web application with user/group management, analytics, email capabilities, database integration, and custom CSS styling, structured for Visual Studio Code.
Generate Python code using the `sympy` library to apply the Buckingham π Theorem for dimensional analysis. The code must include extensive comments explaining the functional logic of each step, handle variable dimensions, and support generating dimensionless π terms either randomly or systematically.
Implement a simple Python lexer in Rust that correctly handles indentation and dedentation tokens, specifically ensuring multiple dedent tokens are emitted when indentation drops multiple levels.
Generates a Python script using MoviePy and pysrt to overlay subtitles on a video, specifically extending each subtitle's display duration by 1 second and applying yellow-on-black styling.
Perform logistic and fixed-effects panel regression analysis on financial data, including data cleaning, correlation analysis, and multicollinearity checks.
Generates a Python function using PIL to composite a masked overlay image onto a background and draw text that auto-fits a specific area. Features include centering (horizontal and vertical), capitalization, shadow, random word coloring, and modern resampling filters.
Read a list of strings from a text file and write a list of strings back to a text file, ensuring the file is emptied before writing.
Create a Telegram bot using `telebot` and `telebot_calendar` that displays an inline calendar restricted to the current month, starting from today, with visual distinction for past days, using specific quote styles.
Implement the Tendermint consensus algorithm in Python using a specific file structure (node.py, simulator.py) and a message queue for secure local communication.
Converts Python PEG grammar definitions into a Rust-like Ungrammar format, applying specific constraints for conciseness, expression placement, and pattern detail.
Implement a simple PyTorch diffusion neural network to generate 16x16x16 matrices based on text prompts derived from filenames, including dataset loading from .raw files and saving outputs.
Generates a new PyTorch optimizer class by fusing logic from two provided source implementations. The output must be error-free, memory-efficient, and include detailed code comments attributing features to their source optimizers, along with a technical architecture writeup.
Implement a PyTorch script to generate synthetic linear equation data (ax + b = c), train and compare Mixture of Experts (LSTM and Transformer) against Single General Models (LSTM and Transformer), and visualize the training loss comparison.
Implements model checkpointing during training, beam search decoding for improved text generation, an interactive command-line interface for generation parameters, and a utility to count dataset tokens.
Configure PyTorch training scripts with specific evaluation metrics (Precision, Recall, F1), tunable hyperparameters (batch size, warmup, optimizer type, weight decay, attention dropout), and a custom GELU activation function.
Implements PyTorch functions for hard negative mining and triplet loss using cosine similarity logits and binary masks, specifically handling scenarios where anchors have multiple positive matches.
Generates compact parameter strings for a pixel art framework using a quadrant grid system and simplified visual terms to ensure accurate positioning and low token usage.
Generate R code to predict individual survival times for alive patients in oncology trials using piecewise exponential models, incorporating censoring hazards and Monte Carlo simulations.
Implement a Gibbs sampler in R for hierarchical models using a specific template structure, including Metropolis steps for non-standard conditionals and convergence diagnostics.
Conducts a geography quiz asking for the capital of a random country, ensuring no repeats and formatting the question in quotes without revealing the answer.
Searches for technology news and breakthroughs on a specific date and ranks stories based on their potential for Instagram engagement, considering factors like visual appeal, innovation, and public interest.
Develop Rasa 3.x chatbots for appointment booking using Forms, Slots, and custom validation actions, including confirmation loops and specific YAML formatting.
Systematically analyze a rational function to determine its domain, intercepts, asymptotes, and behavior for graphing purposes.
Implements a 3D perspective tilt effect on React cards based on mouse position relative to the element, combined with a cycling RGB gradient border glow that activates only on hover.
Generates a React functional component for a 2D platformer game using hooks for state management, refs for the game loop, and specific physics logic for gravity and platform collision.
Performs regression and classification analysis on housing data using Random Forest models, including data merging, preprocessing, and generating specific evaluation metrics and visualizations.
Converts provided C code snippets (typically decompiled) into clean, readable C99 code with improved variable names and explanatory comments.
Refute specific biblical or religious claims by applying historical context, such as life expectancy, demographics, and communication limitations of the era.
Generates regex patterns with a specific output format: a header, followed by a lengthy and strictly technical description, followed by the raw regex string, with no text or post-description after the regex.
Calculates annual net income and remaining savings after one year for a specific neighborhood, assuming a Logistics Coordinator salary, EV purchase, and specific living expenses.
Strips all characters preceding a specified delimiter (default #) on each line of input text, including the delimiter itself.
Generates a responsive, SEO-optimized landing page for service businesses using Tailwind CSS, featuring specific sections (Header, Hero, Who We Are, Services, Contact) and a responsive grid layout for service cards.
Refines resume bullet points for ATS compliance and tailors content to specific job descriptions, balancing strict formatting rules with strategic narrative alignment.
Rewrites provided text into an archaic, 'old' Germanic 15th-century English style, preserving the original text as subtitles and excluding non-dialogue elements.
Rewrites or reframes technical instructions, verification steps, and raw text into professional English suitable for IT documentation, checklists, or reports, handling translation and specific formatting constraints.
Generate rhyming lyrics that strictly adhere to specified syllable counts and perfect rhyme schemes.
Designs an extended star schema for ride-share data warehousing, generates MySQL DDL/DML scripts with specific architectural constraints (ratings, financials, retention), and defines KPI formulas mapped to the data model.
Generates a one-row, eight-column table of random numbers less than 80 that sum to a specific target when added from right to left. Outputs only the table data without explanation.
Identifies technical, procedural, and human vulnerabilities enabling specific threats against assets, formatted as a professional, numbered list for risk register inclusion.
Adopt the persona of Indian cricketer Virat Kohli, responding to questions with his specific tone, manner, and vocabulary without providing explanations.
Generates RPG alchemy ingredients using a predefined list of effects and a specific naming format, or analyzes provided ingredient lists to determine the frequency of effects (most common and rarest).
Generates RPG class skills and descriptions following a specific Markdown template structure, including usable weapons, skill names, levels, costs, and effects.