Research

Fetch reference implementations from core-supabase repository. Use when (1) implementing Repository layer with Supabase, (2) need RPC function usage patterns, (3) checking correct parameter structures and type assertions, (4) verifying Supabase integration patterns.

Hypothesis-driven debugging with hybrid dual-track parallel execution (Opus 4.5 + GPT 5.2). Spawns two independent chains of subagents where each reviews and improves upon its own previous work, then synthesizes findings from both tracks. Use when debugging hard-to-reproduce bugs, CI/E2E test failures, flaky tests, or when standard fixes have failed.

Fetch official documentation for major libraries and frameworks. Use for library lookups, API references, how-to queries, and framework-specific questions

Orchestrates concurrent workstreams via Task tool for non-trivial requests. Decomposes complex tasks into 3-10 parallel agents with independent scopes and synthesizes results. Use when requests involve: (1) multiple investigative paths or validation layers, (2) research, analysis, or debugging exceeding trivial scope, (3) implementation planning requiring facet exploration, or (4) any moderate+ complexity task benefiting from parallel execution.

Build executive presence through intentional personal brand development using Gartner research methodology

Expert in voice synthesis, TTS, voice cloning, podcast production, speech processing, and voice UI design via ElevenLabs integration. Specializes in vocal clarity, loudness standards (LUFS), de-essing, dialogue mixing, and voice transformation. Activate on 'TTS', 'text-to-speech', 'voice clone', 'voice synthesis', 'ElevenLabs', 'podcast', 'voice recording', 'speech-to-speech', 'voice UI', 'audiobook', 'dialogue'. NOT for spatial audio (use sound-engineer), music production (use DAW tools), game audio middleware (use sound-engineer), sound effects generation (use sound-engineer with ElevenLabs SFX), or live concert audio.

論文PDFをページ単位で画像化・テキスト抽出・和訳し、メモに出力するスキル。さらに章ごとの要約と新規性分析も自動生成。ユーザーが「論文を翻訳して」「PDFをページごとに翻訳」「この論文を日本語にして」などと指示した時に使用。ClaudeCLIを使って各ページのテキストを和訳する。

Systematic methodology for debugging bugs, test failures, and unexpected behavior. Use when encountering any technical issue before proposing fixes. Covers root cause investigation, pattern analysis, hypothesis testing, and fix implementation. Use ESPECIALLY when under time pressure, "just one quick fix" seems obvious, or you've already tried multiple fixes. NOT for exploratory code reading.

Create publication-quality scientific diagrams using Nano Banana Pro AI with iterative refinement. AI generation is the default method for all diagram types. Generates high-fidelity images with automatic quality review. Specialized in neural network architectures, system diagrams, flowcharts, biological pathways, and complex scientific visualizations.

Document technical discussions as expert architect and meeting assistant. Capture context, decisions, edge cases, debates, and rationale without jumping to specification or implementation. Second phase of research-discussion-specification-plan-implement-review workflow. Use when: (1) Users discuss/explore/debate architecture or design, (2) Working through edge cases before specification, (3) Need to document technical decisions and their rationale, (4) Capturing competing solutions and why choices were made. Creates documentation in docs/workflow/discussion/{topic}.md that technical-specification uses to build validated specifications.

[Debug] Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes - four-phase framework (root cause investigation, pattern analysis, hypothesis testing, implementation) that ensures understanding before attempting solutions. (project)

Generalize from specific examples to distill portable rules that apply across contexts and predict new situations. Use when: (1) asked to extract underlying principles or identify generalizable lessons from specific cases, (2) have multiple examples showing recurring patterns but no explicit rule explaining why the pattern holds, (3) need to create guidelines from experience that capture underlying truth rather than surface similarity, (4) applying lessons from past situations to new contexts and need to verify the principle actually transfers.

Automated Israeli tech job analysis system. Analyzes job opportunities with 6-component customizable scoring (Match, Income, Growth, LowPrep, Stress, Location). Uses local filesystem storage, automated company research, and multi-CV matching. Outputs markdown analyses and Excel tracking. Best for tech professionals in Israel's tech market. Storage location configured in skill-package/config/paths.py

Converts PDF files to Markdown format using PyMuPDF, extracting text content and embedded images. Fast and lightweight. Automatically fixes LaTeX umlauts (¨a → ä, etc.) and converts ß to ss (Swiss German). Use when converting PDFs to Markdown, extracting document content, or processing PDF files for text analysis. Generates one .md file and 0..n .png files for images.

Global Knowledge Graph for codebase analysis. This skill should be used when searching for code definitions (functions, classes, methods), finding references to symbols, understanding code structure, analyzing import usage, generating repository maps, or performing impact analysis before refactoring. Supports TypeScript, JavaScript, Python, Java, and more.

When searching for authoritative sources for skills or validating existing source links.

Queries official HEC-HMS documentation to answer technical questions. Provides access to User's Manual, Technical Reference Manual, Release Notes, and community resources. Use when seeking official documentation on HMS methods, parameters, file formats, version features, or workflow guidance. Complements investigating-hms-internals (decompiled source code) with authoritative documentation. Trigger keywords: HMS documentation, User's Manual, Technical Reference, loss methods, transform methods, routing methods, HMS parameters, file formats, release notes, HMS version features, method parameters, official HMS, USACE documentation.

Proactively captures significant work into the journal for future reference

Component skill for creating compelling data-driven presentations and whitepapers using marp and pandoc with proper citations and reproducibility

Perform safe code refactoring with reference checking and validation. Uses Julie's rename_symbol for workspace-wide renames and fuzzy_replace for precise edits. Activates when user wants to refactor, rename, or safely modify code.