全部能力

Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text…

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or…

Design and implement beautiful, fluid Liquid Glass interfaces in Expo React Native apps. Covers four paths: (1) expo-glass-effect for UIKit-backed glass surfaces, (2) @expo/ui SwiftUI integration for native SwiftUI glass modifiers and advanced transitions, (3) Expo Router unstable native tabs for system Liquid Glass tab bars, and (4) @callstack/liquid-glass as a third-party alternative. Use when tasks mention "liquid glass", "glass effect", "frosted/translucent UI", "iOS 26 design", "native tabs", "expo-ui", "SwiftUI in Expo", or when shipping Apple-style glass with robust fallbacks, accessibility checks, HIG-aware design decisions (Foundations, Patterns, Components, Inputs), and cross-platform degradation.

Master email marketing from subject lines to sequences. Templates for welcome emails, nurture campaigns, sales emails, and newsletters that get opened, read,…

State-space model with O(n) complexity vs Transformers' O(n²). 5× faster inference, million-token sequences, no KV cache. Selective SSM with hardware-aware…

Visualize training metrics, debug models with histograms, compare experiments, visualize model graphs, and profile performance with TensorBoard - Google's ML…

RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux…

Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT. Use when compressing models without retraining, achieving 50%…

Calculate ROI for marketing campaigns, investments, and business decisions. Includes break-even analysis, payback period, and comparative ROI.

Atomic commits, PR size limits, commit thresholds, stacked PRs

Transform raw data from CSVs, Google Sheets, or databases into executive-ready reports with visualizations, key metrics, trend analysis, and actionable…

Generate images and videos using fal.ai AI models

Use when implementing 'CloudKit sync', 'CKSyncEngine', 'CKRecord', 'CKDatabase', 'SwiftData CloudKit', 'shared database', 'public database', 'CloudKit zones',…

End-to-end, API, and responsive testing for web applications using Playwright with TypeScript. Use when asked to write, run, debug, or maintain Playwright…

在 Linux 主机上安装和配置 OpenClaw，集成钉钉、飞书、Discord 及更多渠道，使用 Bailian/DashScope 模型。在配置…时使用。

Create category-aware, AEO-optimized blog posts for Lightfast. Use when writing technology deep-dives, company announcements, or product launches.

Use when writing, running, or debugging XCUITests. Covers element queries, waiting strategies, accessibility identifiers, test plans, and CI/CD test execution…

Configure Claude Code authentication for Cyrus — API key, OAuth token, or third-party provider.

Configure Aspire AppHost to emit explicit app config via environment variables; keep app code free of Aspire clients and service discovery.

Use when implementing ANY computer vision feature - image analysis, object detection, pose detection, person segmentation, subject lifting, hand/body pose…