論文

Hypothesis validation and experiment design. Use when mapping assumptions, designing experiments, or testing MVPs.

Journaling partner that helps extract deeper meaning from daily logs. Use when the user wants to review their day, process logs into journal entries, or mentions "daily review".

Use when analyzing another branch's iteration journals to extract findings, decisions, and insights from divergent work

arXiv 論文の内容を取得・要約するスキル。URL が arxiv.org/abs/{論文ID} 形式の場合に使用。PDF をダウンロードして Read ツールで読み取る。

Build slide decks and presentations for research talks. Use this for making PowerPoint slides, conference presentations, seminar talks, research presentations, thesis defense slides, or any scientific talk. Provides slide structure, design templates, timing guidance, and visual validation. Works with PowerPoint and LaTeX Beamer.

Statistical analysis methods, hypothesis testing, and probability for data analytics

Identify unifying concepts and common threads that run through multiple observations or ideas, useful for making sense of diverse feedback, understanding what really matters, or finding the signal in noise. Use when: (1) explicit request to identify patterns or themes that unify diverse observations, (2) numerous qualitative observations need analysis for what recurs across them, (3) deciding priorities requires understanding what concerns keep appearing across stakeholders, (4) multiple sources describe situations using different terms but may share underlying patterns.

Use when user wants to assess progress toward an open-ended goal by reading and summarizing all iteration journals

Convert dense source-grounded content into a two-speaker dramatized script with persona roles, prosody cues, and interruptions. Use for audio-friendly overviews that stay faithful to cited material.

Meta-cognitive reasoning specialist for evidence-based analysis, hypothesis testing, and cognitive failure prevention. Use when conducting reviews, making assessments, debugging complex issues, or any task requiring rigorous analytical reasoning. Prevents premature conclusions, assumption-based errors, and pattern matching without verification.

Performs comprehensive, multi-layered research on any topic with structured analysis and synthesis of information from multiple sources. Use when the user needs thorough investigation, market research, technical deep-dives, due diligence, or comprehensive analysis on any subject.

Use when analyzing critical documents, specifications, or large files (>3000 lines), before any synthesis or conclusions - enforces complete line-by-line reading with quantitative verification to prevent skimming that leads to incomplete understanding

Hypothesis-driven debugging with systematic investigation. Identifies root causes, creates minimal reproductions, and implements fixes. Use when investigating errors, bugs, crashes, or unexpected behavior.

Process voice memos with AI transcription and analysis. Multi-language support (EN, HE), speaker identification, action item extraction with priorities, smart summaries, and auto-categorization (meeting, journal, brainstorm, interview). Triggers - "process voice memos", "transcribe", "analyze memo", "show transcripts", "voice inbox", "extract action items", "meeting notes", "transcribe audio".

Generate publication-ready scientific figures for research papers.Use this skill when:- Creating any figures for manuscripts (Nature, Science, Cell quality)- Generating data plots from CSV/Excel/JSON datasets- Creating conceptual diagrams, flowcharts, or schematics- Composing multi-panel figures with letter labels (A, B, C, D)- Visualizing architectures, timelines, or workflowsThis skill handles THREE types of figures:1. DATA PLOTS - Line, scatter, bar, violin, heatmap, etc. from tabular data2. SCHEMATICS - AI-generated conceptual diagrams and illustrations3. COMPOSITIONS - Multi-panel figures combining plots and/or schematicsPrerequisites:- For plots: pandas, matplotlib, seaborn- For schematics: OPENROUTER_AI_API_KEY environment variable- For composition: matplotlib or PIL

Automatically fetch and summarize the latest AI news, research, and industry developments. Use when users request: (1) Daily AI news updates, (2) Latest AI technology developments, (3) Recent AI research papers or breakthroughs, (4) AI industry trends and market news, (5) Specific AI company announcements, or (6) Automated daily AI briefings. Supports customizable search queries, multi-source aggregation, and formatted output in various styles.

Transform raw research into actionable insights that inform design decisions. Use during Define phase after completing research.

Use when user says "synthesize goals", "show goals", "what are my goals", "update goals", "show non-goals", "what am I building", "project direction", or asks about the overall direction or intent of the project - generates two markdown files summarizing the user's goals based on approved story nodes (what the goals ARE) and rejected story nodes with notes (what the non-goals ARE).

Master user research with interviews, surveys, personas, journey mapping, and research synthesis for user-centered design.

Develop stronger ideas by examining thesis and antithesis to reach synthesis that transcends the original opposition. Use when: (1) asked to explore opposing viewpoints or find synthesis between conflicting positions, (2) opposing approaches both have merit or contrary positions present valid reasoning, (3) debates become polarized and people talk past each other, (4) framing forces binary choice between inadequate options or discussions need reframing.