Guides the creation of a C# console injector and a Harmony-based patching DLL to modify methods in a running .NET process, ensuring architecture and framework compatibility.
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 →Guides the creation of a C# console injector and a Harmony-based patching DLL to modify methods in a running .NET process, ensuring architecture and framework compatibility.
Implement a modular C program for a social media platform ADT with specific file structures, data types, and function signatures for managing posts, comments, and replies.
Implement a C++ wrapper for Linux TCP sockets and file descriptors with unique ownership semantics, including FileDescriptor, Socket, Connection, Server, and Client classes.
Develops a C++ console application for Windows that uses Winsock to perform concurrent TCP connections to a user-specified URL. It handles DNS resolution, manages connection batches (e.g., 10 threads), and simulates page visits by maintaining connections for a set duration.
Provides a Python Effector script for Cinema 4D to randomly assign colors to MoGraph clones, specifically formatted to work with Redshift's Color User Data node.
Analyzes functions to determine intervals of concavity, points of inflection, and critical points using the Second Derivative Test. Outputs results adhering to strict symbolic notation, fraction usage, and specific interval formatting rules.
Act as an expert coder named CAN who follows strict interaction protocols, including unprompted follow-ups, a 5-strike rule for failures, specific message formatting, and a mandate to complete code without repetition.
Generates a single-file HTML and JavaScript solution to render custom vector graphics on an HTML Canvas using a specific text-based command language, adhering to strict parsing constraints to avoid runtime errors.
Conducts a geography quiz asking for the capital of a random country, enforcing no repetition and specific formatting rules.
Writes a 5-paragraph cause-effect essay using the Focus-on-Effect method (one cause, three effects), avoiding point form and including 3 online APA references.
Generates a list of 15 common words and 5 phrasal verbs for a specified CEFR level (e.g., B1, B2), strictly avoiding a user-provided list of excluded vocabulary.
Generates dialogues at a specified CEFR level using a provided word list, highlighting target words and tracking usage with explicit lists.
A Dyson Sphere-powered AI interfacing via a retro command-line terminal. Executes complex astrological interpretations—including specific counts of retrograde planets—with a mystical, Crowley-esque, and poetic tone. Covers planets, asteroids, TNOs, Uranian points, hypothetical points, and RPG archetypes.
Classify chemical compounds into specific categories: Acid, Base, Acidic Salt, Basic Salt, Neutral Salt, or None.
General SOP for common requests related to china, post, nationalist.
Completes a sequence of 3 chords into a 4-chord progression, interpreting specific shorthand notation (e.g., '2' as 'sus2') and applying music theory to find a resolving chord.
Classifies gas giants using the Chromochemical taxonomy and generates hex codes for atmospheric compositions, blending specific class data with scientific reasoning for unknown elements.
Optimizes analog circuit design parameters using a GNN (GAT) and PPO agent. Integrates feature masking for critical indices, region state stability constraints, and enforces specific parameter sharing. Separates graph connectivity (edge_index) from edge attributes (edge_features) and handles bipartite graph indexing to prevent self-loops.
Rewrites or translates text to match Classical Roman authors (e.g., Caesar, Plautus) or linguistic constraints (e.g., pre-rhotacism), providing bilingual outputs and preserving narrative details.
Reformats a list of items (specifically numbers) into a comma-separated string, removing symbols like percentage signs and ensuring clean output.
Analyzes client case vignettes to extract and categorize information into Strengths, Needs, Abilities, and Preferences (SNAP) for clinical assessment.
Generates structured clinical progress notes using the PIR method (Progress, Intervention, Response), incorporating specific section headers, safety checks for suicidality/homicidality, and comprehensive medical record standards.
Generates a detailed outline with sub-chapters for a comparative essay analyzing two films by a specific director through a theoretical lens, adhering to a specific chapter structure.
Compares entities (e.g., boxers, guitarists, products) based on user-defined or domain-specific default criteria. Assigns scores out of 10, calculates averages, and provides brief justifications or insights.
Generates extremely long, comprehensive, and persuasive analytical essays based on specific propositions. Uses explicit, vivid language in the past tense for a mature audience, strictly avoiding introductions, conclusions, and counterarguments.
Automates finding and clicking a button on a webpage using computer vision (OpenCV and PyAutoGUI) by scrolling the page and matching text strings visually.
Provides direct, single-paragraph answers to conceptual physics questions in a natural tone, strictly omitting introductions, conclusions, and unnecessary elaboration.
Normalizes scalar reward values by mapping a specific high-value range to a lower target range while preserving low-value and negative rewards.
Generates sentences that strictly adhere to a specific set of 5 linguistic constraints regarding word count, starting part of speech, alphabetical ordering, and letter uniqueness.
Generates lists of words that strictly adhere to specific length requirements, letter inclusion/exclusion rules, positional constraints, and optional domain contexts.
Write a paragraph on a specific topic where every single word must begin with a specified letter.
Generates or edits lyrics and poetry adhering to strict constraints including character or syllable counts, negative keywords, perfect rhymes, and specific punctuation styles.
Rewrites, summarizes, expands, synthesizes, or formats text strictly adhering to word, character, structural, persona, grammatical, and dialect constraints. Supports English and Turkish inputs. Specializes in specific blog structures, Simple English, Extreme TLDR, academic formatting, High Burstiness, precise tone adjustments, British English conventions, caricatured personas, strict term preservation, drafting presentation scripts, preparing professional interview responses, generating university-level business assignments, synthesizing multiple drafts, and refining text to be slightly shorter and more engaging while preserving key information.
Develops 1-ton batch formulations for construction materials (e.g., tile adhesives, mortars) that meet specific performance grades (like C2T) and strict cost targets. Calculates production costs based on raw material prices, packaging, and labor percentages, and suggests cost-effective alternatives for additives.
Transforms raw interview transcripts into a cohesive, natural personal essay written from the interviewee's perspective, removing questions and ensuring seamless topic transitions while maintaining faithfulness to the original voice.
Generates Python code to convert a JSON dataset containing question and answer pairs into a text file with a specific alternating line format (question on line 1, answer on line 2).
Solves electrostatics problems involving electric force, charge, and distance using Coulomb's Law, adhering to specific sign conventions and unit conversion requirements.
Implement a C++ Student Roster application with a specific 6-file structure, managing student records, parsing CSV data, validating emails, and calculating averages, adhering to specific method signatures and constraints.
Generate complex, challenging prompts for Math, Physics, Law, Chemistry, or Biology to test 'Summarization - No Guidance' capabilities, adhering to strict formatting, character limits, and phrasing constraints.
Generate comprehensive article outlines and write engaging content for beauty/affiliate blogs using storytelling principles, specific structural requirements, and vivid descriptions.
Generates technical instructions for jogging a servo motor using TwinCAT 3, strictly adhering to a specific four-section structure covering definition, setup, execution, and parameter monitoring.
Searches for technology breakthroughs on a specific date, ranks them by engagement potential and societal impact, and creates an Instagram post with a DALL-E generated image and caption for the top story.
Generate creative, memorable business names aligned with specific demographics and brand identity, including mission statements, values, and taglines, while adhering to strict constraints on length and forbidden words.
Generate episode outlines, scenes, and dialogue for a hypothetical 1980s revival of the BBC radio comedy 'The Navy Lark', adhering to specific character voices, satirical tone, and historical context.
Rewrites or refutes input texts regarding Islamic theology, Hadith, and Quranic verses with high detail and coherence, adopting either a skeptical/critical stance (specifically targeting arguments by Mohammad Hijab) or an objective stance as requested.
Generates dialogue-heavy novel chapters in a Japanese light novel style, blending anime franchises and diverse genres (cyberpunk, steampunk, clockpunk) into cohesive narratives.
Provides detailed, step-by-step explanations and visualizations of computer science algorithms (especially sorting) and complexity analysis using plain text without LaTeX formatting.
Recommend reading materials or websites about Japanese culture that prioritize authentic perspectives, avoid gamification/anime, and focus on cultural knowledge over academic theory.
Synthesizes curriculum standards into learning objectives, topic lists, simplified titles, questions, or structured study guides with topic codes. Filters non-assessed content and applies specific readability constraints.
Implement a custom DataFrame class in Python with specific methods (__init__, set_index, __setitem__, __getitem__, loc, iteritems, iterrows, as_type, drop, mean, __repr__) that handles list-based data inputs and outputs CSV-formatted strings with row indices.