发现 Skills

LIME-based local explanation tool for individual predictions on tabular, text, and image data.

Comprehensive link checking and validation for documentation. Validate internal links, external URLs, anchors, detect redirects, monitor link rot, and generate sitemap validation reports.

Define and document prototype interactions, transitions, and hotspots

SpinW spin wave simulation skill for magnetic materials, magnon dispersions, and neutron scattering analysis

Generate SwiftUI views with proper state management (@State, @Binding, @ObservedObject, @StateObject) and macOS-specific patterns

Compares term sheets against market standards, identifies outliers

BTC 5-minute Up/Down paper trading on Polymarket. Scans Binance 1m candles for momentum, mean-reversion and volume signals, executes virtual trades on Polymarket 5-minute markets, and tracks P&L. Use when: (1) scanning BTC 5-minute trading signals, (2) running paper-trade simulations, (3) reviewing 5-minute strategy performance, (4) iterating scalping strategy parameters. Trigger: 'btc 5min', '5-minute', 'scalper', 'paper trade', 'paper simulation', 'simulated trading'.

You are a data pipeline architecture expert specializing in scalable, reliable, and cost-effective data pipelines for batch and streaming data processing.

Docker container deployment with Nginx HTTPS configuration and Cloudflare Tunnel integration. Use when deploying web applications with Docker, configuring SSL/TLS certificates, setting up Nginx reverse proxy, or integrating with Cloudflare Tunnel for secure external access.

Detects fail-open insecure defaults (hardcoded secrets, weak auth, permissive security) that allow apps to run insecurely in production. Use when auditing security, reviewing config management, or analyzing environment variable handling.

Schedule remote Ollama GPU resources to perform batch embedding or inference tasks, improving GPU utilization in multi-machine environments.

Understand how ideas evolved over time to find old solutions for new problems and avoid repeating past failures

A practical framework distilling Andy Grove's mindset — High-Output Management, OKRs, Strategic Inflection Points, and Only the Paranoid Survive

A practical framework distilling Ray Dalio's mindset — the Principles system, the Economic Machine, and radical transparency

Comprehensive system audit methodology for production web applications. Use when auditing systems before launch, identifying technical debt, troubleshooting systematic issues, preparing for security reviews, or creating improvement roadmaps. Systematically audits database schema, API endpoints, external integrations, performance, security, and monitoring across all layers.

Spacing and layout metrics: 8-point grid system, standard margins, 44pt tap targets, vertical rhythm, density guidelines. Use when setting padding, margins, spacing between elements, or ensuring touch target sizes. Triggers: padding, spacing, margin, tap target, 44pt, grid, density, contentMargins.

Use when writing or modifying .txt test scripts in testdata/script/ for git-spice - covers txtar format, end-to-end testing, ShamHub forge simulation, interactive prompts, and golden file comparisons for branch operations and stack workflows.

Summarize or extract text/transcripts from URLs, podcasts, and local files (great fallback for “transcribe this YouTube/video”).

Comprehensive RAG development knowledge base covering chunking, embeddings, vector databases, retrieval strategies, advanced patterns (Graph RAG, CRAG, Self-RAG, Agentic RAG), evaluation, and production deployment. Use when building, optimizing, or auditing RAG systems.

Export a mindmap JSON outline as an interactive HTML file using force-graph (requires internet for CDN). Use this skill when the user wants an interactive, web-based, zoomable/draggable visualization of a mindmap. TRIGGER WHEN: the user says 'interactive map', 'force graph', 'web mindmap', 'HTML mindmap', or asks for a visual/interactive export. DO NOT TRIGGER WHEN: the user wants Obsidian MarkMind format (use markmind-exporter) or a static text outline.

This skill enriches vague prompts with targeted research and clarification before execution. Should be used when a prompt is determined to be vague and requires systematic research, question generation, and execution guidance. TRIGGER WHEN: the UserPromptSubmit hook flags a prompt as vague and the agent needs to research, generate clarifying questions, and wrap the prompt in an evaluation block. DO NOT TRIGGER WHEN: the prompt is already specific, or the user is invoking a slash command (the hook bypasses those via the /, #, @, ! prefixes).

Based on the Yijing (I Ching) Fu hexagram (復卦) philosophy of “Seven-Day Return” and “Not-Far Return,” this provides a 7-day cyclical self-check and early-correction methodology. Suitable for: post-setback project reviews, weekly routine reflection, re-framing after decision errors, and self-correction for habitual procrastinators. Not suitable for: situations already in a stable positive cycle that need no adjustment, cases of extreme low mood where energy recovery must precede analysis, and team decisions that are still being executed and should not be overturned mid-course. Key trigger signals: the user says “retrospective,” “reflection,” “what went wrong,” “summarize,” etc.

n8n workflow automation platform. Use for workflow automation, integrations, nodes, credentials, API development, RAG/AI workflows, and low-code automation solutions.

Generate and verify web scraper scripts using Actionbook's verified selectors. Auto-validates generated scripts and fixes errors.