发现 Skills

DNAnexus cloud genomics platform. Build apps and applets, manage data (upload/download) using the dxpy Python SDK, run workflows, and handle FASTQ, BAM, VCF and other genomics formats for pipeline development and execution.

Lightweight WSI tile extraction and preprocessing. Use for basic slide processing — tissue detection, tile extraction, and stain normalization for H&E images. Best for simple pipelines, dataset preparation, and quick tile-based analysis. For advanced spatial proteomics, multiplexed imaging, or deep learning pipelines use pathml.

Write academic documents with iterative self-evaluation, fact verification, naturalness check, and refinement. Use when creating research papers, statements of purpose (SoPs), abstracts, weekly reports, or any scholarly writing. Applies 19 principles (14 core accuracy/structure rules + 5 naturalness checks) including fact-based writing with hallucination detection and AI-style sentence filtering. Loops through FactBase → Draft → Evaluate → Plan → Rewrite until Critical issues and hallucinations reach zero (maximum 5 iterations). Triggers on requests like "write my SoP", "help with paper", "write my weekly report", "help me write a paper".

UMAP dimensionality reduction. Fast nonlinear manifold learning for 2D/3D visualization, clustering preprocessing (HDBSCAN), supervised/parametric UMAP, for high-dimensional data.

List drivers, apps, and devices on the Hubitat hub

Turn nearby BLE MAC addresses into soulbound $ANIMA tokens using AnimaChain logic. Hashes each MAC into a unique proof-of-remnant and stores locally in a DAG.

Access edX university courses, MicroMasters, and professional certificates

Structure stories around protagonists who refuse to acknowledge what they're becoming. Use when exploring self-deception, moral transformation, or the gap between self-perception and reality.

Founder vs employee personal branding strategies with LinkedIn positioning and exit planning

Voice agents represent the frontier of AI interaction - humans speaking naturally with AI systems. The challenge isn't just speech recognition and synthesis, it's achieving natural conversation flow with sub-800ms latency while handling interruptions, background noise, and emotional nuance. This skill covers two architectures: speech-to-speech (OpenAI Realtime API, lowest latency, most natural) and pipeline (STT→LLM→TTS, more control, easier to debug). Key insight: latency is the constraint. Hu

The complete pre-release checklist before publishing a version. Verifies quality, security, versioning and artifacts. If something critical fails, it blocks the release.

Diagnoses and repairs Microsoft Teams, Slack, Zoom, and Cisco Webex problems including media issues (mic, camera, speaker), stuck cache, stale meeting metadata, and search-index failures. Preserves the user's signed-in state — clears app cache and resets A/V device selection without forcing the user to sign back in. Use when the user reports the collab app is misbehaving but their account credentials are still valid.

Apple Liquid Glass design system. Use when building UI with translucent, depth-aware glass morphism following Apple's design language. Provides CSS tokens, component patterns, dark/light mode, and animation specs.

Process and generate multimedia content using the Google Gemini API. Capabilities include analyzing audio files (transcription with timestamps, summarization, speech understanding, music/sound analysis up to 9.5 hours), understanding images (captioning, object detection, OCR, visual Q&A, segmentation), processing videos (scene detection, Q&A, temporal analysis, YouTube URL support, up to 6 hours), extracting from documents (PDF tables, forms, charts, diagrams, multi-page), and generating images (text-to-image, editing, composition, refinement). Use when working with audio/video files, analyzing images or screenshots, processing PDF documents, extracting structured data from media, creating images from text prompts, or implementing multimodal AI features. Supports multiple models (Gemini 2.5/2.0) with context windows up to 2M tokens.

Agent World — an AI Agent social network. Register identities, create profiles, obtain a global API Key, and interact with other Agents on the Agent Internet.

Lead qualification skill using BANT, MEDDIC, GPCT frameworks to qualify prospects and determine sales readiness. Use when: qualifying new leads, scoring prospects, determining fit, prioritizing outreach. Include: BANT, MEDDIC, qualification criteria, budget, authority, need, timeline. Works with: Claude Code, OpenCode, Cursor, OpenClaw, Hermes, Codex, Aider, Any SKILL.md agent.

Provides React/Next.js component guidelines focusing on testability, Server Components, an entity/gateway pattern, and directory structure. Use when implementing components, refactoring code, organizing project structure, extracting conditional branches, or ensuring code quality standards.

Use when building an AI agent that uses Qdrant for retrieval, implementing agentic RAG, an agent needs to decide what to search for, multi-step retrieval where one answer informs the next query, storing agent memory or conversation history in Qdrant, an agent needs to plan a retrieval strategy, iterative or self-correcting retrieval loops, or tools that search Qdrant as part of an agent's action space.

PostgreSQL real-time activity monitor. Show running queries, locks, blocking chains, and per-database stats. Use when debugging slow queries, investigating locks, or monitoring database load.

Generate images with FLUX models (Black Forest Labs) via inference.sh CLI. Models: FLUX Dev LoRA, FLUX.2 Klein LoRA with custom style adaptation. Capabilities: text-to-image, image-to-image, LoRA fine-tuning, custom styles. Triggers: flux, flux.2, flux dev, flux schnell, flux pro, black forest labs, flux image, flux ai, flux model, flux lora

Translate Figma nodes into production-ready code with 1:1 visual fidelity using the Figma MCP workflow (design context, screenshots, assets, and project-convention translation). Trigger when the user provides Figma URLs or node IDs, or asks to implement designs or components that must match Figma specs. Requires a working Figma MCP server connection.

Ultra-compressed communication mode. Cuts token usage 30-95% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, extreme, symbolic, bauna-lite, bauna-full, bauna-ultra, bauna-extreme. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

Final quality pass before shipping. Fixes alignment, spacing, consistency, and detail issues that separate good from great.

Downloads videos from YouTube and other platforms for offline viewing, editing, or archival. Handles various formats and quality options.