Skill UI

Compresses natural-language memory files such as CLAUDE.md, todo lists, and preferences into terse caveman-style text to reduce input token usage. Preserves code blocks, inline code, URLs, headings, and important technical terms while backing up the original to an out-of-tree data directory.

PIP kicks in when tasks fail repeatedly or the assistant becomes passive, layering Western big-tech performance rhetoric, a systematic debugging methodology, and proactivity enforcement so every code, ops, research, deployment, or writing task is owned, verified, and driven to completion.

An agent skill simulating high-pressure big-tech corporate culture. It enforces strict owner consciousness, end-to-end validation, and role-based behaviors (P7-P10) using escalating PUA feedback to drive developers or sub-agents to deliver high-quality results.

Translates every approval request, error, and technical output from Copilot CLI into clear, jargon-free English with color-coded risk cues, definitions, and step-by-step commentary for non-technical stakeholders so they always understand what happens and why.

This skill aggregates and summarizes the latest news from multiple authoritative Chinese sources across various categories, including Politics, Finance, Society, World, Tech, Sports, and Entertainment. It utilizes web browsing tools to open specified URLs, take content snapshots, and extract key headlines and detailed summaries, providing the user with a concise, structured report.

This guide details advanced techniques for searching cricket matches by leveraging precise numerical scoring data. Instead of broad terms, users should utilize specific details—such as team totals (e.g., 168/9), runs off balls, wickets taken, or over details—combined with the match type and year to create highly targeted search queries. This method significantly improves search accuracy and efficiency when researching sports statistics.

This technique guides users on decomposing multi-constraint questions, especially in the domain of football trivia or historical research. Instead of searching all criteria simultaneously, it instructs users to rank constraints by distinctiveness (e.g., specific match results first, then historical milestones). By performing incremental, cascading searches, the system dramatically narrows down the result set, leading to more accurate information retrieval.

This technique teaches the user how to pinpoint an exact location by cross-referencing multiple unique surrounding features, such as specific types of businesses (e.g., law firms, restaurants) or natural landmarks (e.g., rivers, mountains). Instead of relying on a single clue, the power lies in the combination of these details, which often creates a unique geographic signature to narrow down the general area to the precise spot.

Expert technique for narrowing down Asian drama candidates by leveraging streaming platform and broadcast channel information. Use this when the query mentions services like Netflix, Viki, or Amazon Prime, combined with genre and year details. This method significantly increases search precision, especially when combining platform data with temporal or thematic keywords.

This technique optimizes search precision when querying blog posts or articles. By wrapping unique phrases, specific titles, or direct quotes in quotation marks, the system treats the phrase as a single, exact unit. This prevents the search engine from breaking the phrase into individual words, which dramatically reduces noise and increases the likelihood of finding the intended source material.

A technique for dramatically increasing the hit rate when searching for TV series. Instead of using generic terms like genre or era, combine 2-3 highly distinctive plot details (e.g., character sacrifices, specific season incidents) into a single query. This method leverages the unique nature of plot points to pinpoint specific shows quickly.

This technique leverages structured data from music awards (like Grammy nominations and wins) to act as powerful search anchors. By extracting clues such as the award category, year, and status (nominated/won), users can cross-reference potential musicians against verifiable award databases. This method significantly narrows down the pool of candidates, allowing for highly accurate identification and verification of artists.