Persistent research knowledge base that accumulates papers, ideas, experiments, claims, and their relationships across the entire research lifecycle. Inspired by Karpathy's LLM Wiki pattern. Use when user says "知识库", "research wiki", "add paper", "wiki query", "查知识库", or wants to build/query a persistent field map.
Multi-source ML dataset discovery. Search HuggingFace Hub, OpenML, GitHub, and paper cross-references for datasets relevant to a research task. Use when asked to "find datasets for", "search ML datasets", "what datasets exist for", or "discover training data for".
Use when a quest needs to attach, import, reproduce, repair, verify, compare, or publish a baseline and its metrics.
Generate a structured paper outline from review conclusions and experiment results. Use when user says "写大纲", "paper outline", "plan the paper", "论文规划", or wants to create a paper plan before writing.
Creates implementation plan, writes project code with judge feedback loop, and submits final experiment run. Use after code-survey in both Idea and Plan branches.
Scaffold or draft new MetaMask documentation to editorial standards. Use when creating a new page, writing a first draft, or helping a non-writer meet the documentation team's expectations.
Review MetaMask documentation for editorial compliance (voice, terminology, formatting, content type, frontmatter, workflow). Use before submitting a PR or when asked to audit existing pages.
Reference page structure, templates, and writing patterns for src/content/reference/. For components, see /docs-components. For code examples, see /docs-sandpack.
Analyze the current session and propose improvements to skills. **Proactively invoke this skill** when you notice user corrections after skill usage, or at the end of skill-heavy sessions. Also use when user says "reflect", "improve skill", or "learn from this".
Help users complete Duolingo language lessons. Use when the user asks to do Duolingo, practice a language, complete lessons, or needs guidance navigating the Duolingo app on phone.
Integrate digital health data sources (Apple Health, Fitbit, Oura Ring) and connect to WellAlly.tech knowledge base. Import external health device data, standardize to local format, and recommend relevant WellAlly.tech knowledge base articles based on health data. Support generic CSV/JSON import, provide intelligent article recommendations, and help users better manage personal health data.
Create professional slide decks from topics or documents. Generates structured content with data-driven charts, speaker notes, and complete PPTX files. Applies persuasive storytelling principles (Pyramid Principle, assertion-evidence). Supports multiple formats (Marp, PowerPoint). Use for presentations, pitches, slide decks, or keynotes.
Consolidates redundant documentation while preserving all valuable content. This skill should be used when users want to clean up documentation bloat, merge redundant docs, reduce documentation sprawl, or consolidate multiple files covering the same topic. Triggers include "clean up docs", "consolidate documentation", "too many doc files", "merge these docs", or when documentation exceeds 500 lines across multiple files covering similar topics.
Optimize CLAUDE.md files using progressive disclosure. Goal: Maximize information efficiency, readability, and maintainability. Use when: User wants to optimize CLAUDE.md, information is duplicated across files, or LLM repeatedly fails to follow rules.
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
Create or refine KiraClaw/Codex skills as concise SKILL.md packages. Trigger on: "make a skill", "create a skill", "update this skill", "seed skill", "default skill", or when the user wants reusable agent workflow instructions instead of one-off prompt text.
Build sandboxed applications for secure code execution. Load when building AI code execution, code interpreters, CI/CD systems, interactive dev environments, or executing untrusted code. Covers Sandbox SDK lifecycle, commands, files, code interpreter, and preview URLs. Biases towards retrieval from Cloudflare docs over pre-trained knowledge.
Analyze your Claude Code session logs to improve prompt quality, optimize tool usage, and become a better AI-native engineer.
Generate comprehensive component guideline documentation for SEED Design System. Use when creating or updating design guideline documentation in ./docs/content/docs/components directory. This skill helps create high-quality documentation similar to action-button.mdx.
Generate React component documentation for SEED Design System. Use when creating new React component docs in ./docs/content/react/components or updating existing implementation documentation. Helps document component APIs, props, installation, and usage examples.
