AI Skills
Installable SKILL.md packages that extend AI coding agents with reliable workflows and domain-specific judgment — ranked by installs, sourced from skills.sh.
refined
Modern minimal design system with elegant serif typography and sophisticated color palettes.
esp32-firmware-engineer
ESP32 firmware engineering for ESP-IDF projects. Write, review, and debug embedded C/C++ code involving FreeRTOS tasks/queues/timers, GPIO/I2C/SPI/UART/ADC/PWM peripherals, TWAI/CAN, Wi-Fi/BLE networking, OTA updates, Secure Boot and flash encryption, LVGL display integration, build/flash/monitor workflows, logging, crash analysis, memory/code-size optimization, low-power sleep/wakeup design, on-device USB/serial service terminals, and board bring-up. Use when an agent is asked to implement ESP-IDF firmware features, review embedded changes for correctness or race conditions, investigate boot/runtime failures or Guru Meditation panics, interpret serial logs, fix build/link/flash problems, optimize RAM/flash usage, tune deep sleep/light sleep behavior, harden firmware for production, add a service console/CLI, integrate a display with LVGL, or diagnose hardware-software integration issues on ESP32-class devices.
firebase
Complete backend platform with auth, database, storage, functions, and hosting—optimize for denormalized queries and watch your costs.
multi-search
智能多主题深度研究工具，支持任意材料输入，使用独立研究 Agent进行并行深度检索并生成系统化研究文档。本技能应在用户需要深度研究多个相关课题、进行系统性信息检索、整合多角度分析时使用。
tao-train-segformer
Lightweight transformer-based semantic segmentation with hierarchical feature extraction for real-time inference.
plan-do-check-act
Iterative PDCA cycle for systematic experimentation and continuous improvement
tao-generate-referring-expressions
Generate referring-expression annotations from images with KITTI bounding-box labels using a VLM pipeline.
orpc-contract-first
Guide for implementing oRPC contract-first API patterns in Dify frontend. Trigger when creating or updating contracts in web/contract, wiring router composition, integrating TanStack Query with typed contracts, migrating legacy service calls to oRPC, or deciding whether to call queryOptions directly vs extracting a helper or use-* hook in web/service.
tao-run-automl-deft-pipeline
Run NVIDIA's three-phase AOI training pipeline: AutoML baseline → DEFT loop → AutoML refinement.
tao-run-deft-aoi
Run the full DEFT AOI improvement loop for NVIDIA TAO VisualChangeNet/ChangeNet PCB inspection models.
deepseek-ocr
Vision-language model for OCR, document-to-markdown conversion, and figure parsing with context optical compression.
draw-io
draw.io diagram creation, editing, and review. Use for .drawio XML editing, PNG conversion, layout adjustment, and AWS icon usage.
tao-run-on-brev
Run TAO training/inference containers on NVIDIA Brev GPU instances via the docker contract.
wechat-article-search
Search WeChat public account articles by keyword and retrieve titles, summaries, publish dates, sources, and links.
critique-composition
Critique spatial balance, whitespace, rhythm, and gestalt grouping in screen layouts.
tao-train-centerpose
Train and deploy CenterPose models for 6-DoF object pose estimation via keypoint regression.
auth0-mfa
>
tao-train-pointpillars
PointPillars 3D object detection from LiDAR point clouds for autonomous driving and robotics.
create-pr
Create pull requests using GitHub CLI with proper templates and formatting
writing-job-descriptions
Write effective job descriptions by defining 12-month success and identifying candidate spikes.
clerk-auth
Expert patterns for Clerk auth implementation, middleware,
tao-train-mask-auto-label
Weakly-supervised segmentation from minimal annotations using ViT-MAE backbone.
backend-development
Backend API design, database architecture, microservices patterns, and test-driven development. Use for designing APIs, database schemas, or backend system architecture.
tao-train-mask-auto-encoder
Masked Auto-Encoder for self-supervised vision pretraining and fine-tuning with patch masking and reconstruction.
review-pr
Review an existing GitHub pull request and post inline review comments on its diff. Use when the changes are on an opened PR rather than your local working tree.
tao-train-single-step
Standard single-step train/eval/export workflow for TAO models without AutoML loops.
security-suite
Run composable security analysis.
feishu
Agent skill from mcp-hub.momenta.works.
tao-train-bevfusion
Train BEVFusion models for multi-sensor 3D object detection fusing LiDAR and camera data in bird's-eye-view space.
elevenlabs
Generate AI voiceovers, sound effects, and music using ElevenLabs APIs. Use when creating audio content for videos, podcasts, or games. Triggers include generating voiceovers, narration, dialogue, sound effects from descriptions, background music, soundtrack generation, voice cloning, or any audio synthesis task.