Stereo depth estimation using FoundationStereo. Predicts disparity maps from stereo image pairs for 3D reconstruction. Use when training, evaluating, exporting, or running inference for a TAO FoundationStereo model. Trigger phrases include "train stereo depth", "FoundationStereo", "stereo disparity estimation", "3D reconstruction from stereo".

Run container-backed AutoML / hyperparameter optimization (HPO) for NVIDIA TAO networks using AutoMLRunner. Handles algorithm selection (bayesian, hyperband, asha, bohb, llm, hybrid, autoresearch), WandB experiment tracking, job execution on any TAO SDK platform, result interpretation, and per-rec custom evaluation hooks. Use when the user mentions TAO AutoML, hyperparameter optimization, HPO, automl, automl_settings, AutoMLRunner, tao_automl, bayesian search, hyperband, ASHA, LLM-guided search, autoresearch, or wants to tune train/distill/prune/quantize action parameters for any TAO network. Model actions use the model skill's resolved container image by default; venv training requires an explicit user request. Platform-agnostic — runs on any SDK (Brev, SLURM, Kubernetes, Docker).

Integrate a HuggingFace Computer Vision model into the NVIDIA TAO Toolkit ecosystem (tao-core config, tao-pytorch trainer, tao-deploy TensorRT pipeline). Use when the user asks to "integrate a HuggingFace model into TAO", "add an HF model to TAO Toolkit", "wire a HuggingFace ViT/DETR/ SegFormer into tao-pytorch", "build a TAO trainer + deploy pipeline for an HF CV model", or pastes a HuggingFace model URL/ID and wants it turned into a TAO model. Covers the full 7-phase loop: prerequisites check, HuggingFace inspection and validation, codebase exploration, tao-core configuration and native trainer implementation, ONNX export plus TensorRT deploy integration, packaging and L0 testing, container-based end-to-end validation, and (conditional) accuracy/latency tuning. Supports classification, object detection, semantic / instance / panoptic segmentation, zero-shot detection, and depth estimation.

Use when you need to rebuild the BSP overlay — DT, OOT modules, or kernel — from changes under bsp_sources/. Triggers: build bsp, rebuild dtb, rebuild kernel.

Fine-tune a manifest-backed GR00T or openpi remote Task on compatible LeRobot data. Use for training; do not use for inference-only Tasks or checkpoint rollout.

Serve and visually inspect a converted LeRobot dataset in the browser. Use for videos and state/action timelines; do not use for raw workflow HDF5 or incomplete conversion output.

Launch the interactive Slang fluoroscopy viewport with XPBD catheter physics. Use when asked to open the viewport, teleop a catheter, or demo fluoro navigation.

Render a single DRR fluoroscopy frame from a CT cache or synthetic phantom. Use when asked to render DRR, generate a fluoro image, or smoke-test the Slang renderer.

Use this skill for hands-on DOCA Compress programming on a BlueField DPU, ConnectX NIC, or host with DOCA — enabling compress-deflate, decompress-deflate, decompress-lz4-stream, or decompress-lz4-block tasks on a doca_compress context (the hardware supports DEFLATE both directions plus LZ4 decompress; LZ4 encode is NOT supported), sizing source / destination doca_buf against the per-task cap query, setting mmap permissions, deciding offload vs CPU zlib / zstd, validating with a round-trip smoke, or debugging DOCA_ERROR_* from a Compress call. Trigger on phrasings like "offload this gzip", "decompress incoming network data", "compress task returns INVALID_VALUE on alloc_init", "submitted a task but no completion arrives", or "decompress LZ4 on the BlueField." Refuse and route elsewhere for non-DEFLATE / non-LZ4 algorithms (zstd / Snappy / brotli), LZ4 encode (route to a CPU LZ4 library), pure mmap-to-mmap copies (doca-dma), or DOCA Core lifecycle internals.

Use this skill when building, deploying, evaluating, debugging, or measuring latency for the DeepStream SOP Inference Microservice — a GPU-accelerated FastAPI service that detects whether operators perform assembly-line steps in order via event boundary detection (GEBD) plus VLM classification. Trigger even if the user does not name it: verify operator step sequence, detect missing or out-of-order SOP steps, score factory/work-cell video for procedure compliance, run VLM-based SOP checking on industrial cameras, or call /v1/chat/completions with a file, RTSP, or Basler camera. Also trigger for its internals: SOPVideoProcessor, DeepStream GEBD model (e.g. DDM) via Triton CAPI, nvds_custom_postprocess, Cosmos Reason 1/2 vLLM, SSE streaming, Kafka NvProto/JSON output, Basler/Pylon camera + emulation, Docker compose, chunk-level latency. Do NOT trigger for generic DeepStream pipelines, object detection/tracking, NIM imports, or video summarization.

Use this skill to bring a supported object-detection vision model from HuggingFace or NVIDIA NGC into an NVIDIA DeepStream pipeline with end-to-end automation: ONNX download, SafeTensors export, TRT engine build, custom nvinfer bbox parser, multi-stream benchmark, and PDF report. Object detection models only.

Automates browser interactions for social media management across Instagram, LinkedIn, and X. Handles posting, DMs, connection requests, lead scraping, and monitoring. Use when the user needs to navigate, interact with, or extract data from approved websites.

Create model classes with `fromJson` and `toJson` methods using `dart:convert`. Use when manually mapping JSON keys to class properties for simple data structures.

Configures Flutter Driver for app interaction and converts MCP actions into permanent integration tests. Use when adding integration testing to a project, exploring UI components via MCP, or automating user flows with the integration_test package.

Workflow for fixing package version conflicts. Use this when `pub get` fails due to incompatible package versions.

Resolves deep, structural performance bottlenecks in the Kotlin Android codebase. Use this skill for macro-level improvements like fixing Room N+1 relation queries, optimizing complex multi-measure Compose layouts, implementing Paging 3, refactoring massive UiState classes, or resolving background StateFlow collection battery drains.

Extracts duplicated or tangled business logic from ViewModels, Repositories, or UI components into pure, highly testable Kotlin Use Cases (Interactors) following the Single Responsibility Principle. Use this skill to decouple Android dependencies from business rules, create domain-layer interactors, or reduce ViewModel bloat.

SQL injection testing skill for offensive security assessments and bug bounty hunting. Covers error-based, UNION-based, boolean/time-based blind, out-of-band, second-order, NoSQL, GraphQL, WebSocket, and JSON-operator SQLi. Includes WAF bypass techniques, database-specific exploitation (MySQL, MSSQL, PostgreSQL, Oracle), cloud-native attack paths, ORM CVE tracking, and SQLmap automation. Use when performing web application SQL injection testing, database enumeration, privilege escalation via SQLi, or assessing injection vectors in APIs and modern stacks.

Use when writing Python code with `axllm` for GEPA, Pareto tradeoffs, reflection clients, metric budgets, optimizer state, and artifacts.

Use when the user wants to run SQL — especially analytical SQL — on local files (parquet/csv/json), URLs, S3 paths, or remote databases (Postgres, MySQL, MongoDB, ClickHouse Cloud, Iceberg, Delta Lake) without setting up a server. Provides chDB — embedded ClickHouse SQL in Python with 1000+ functions, Session for stateful multi-step pipelines, parametrized queries, and cross-source joins via `s3()`, `mysql()`, `postgresql()`, `iceberg()`, `deltaLake()`, `remoteSecure()` table functions. TRIGGER when: user wants SQL on parquet/csv/files or across remote analytical sources; uses ClickHouse SQL features (window functions, windowFunnel, geoToH3, JSON path ops, Session, parametrized queries); imports `chdb` or calls `chdb.query()`. SKIP this skill for pandas-style DataFrame method-chaining (use chdb-datastore instead) or ClickHouse server administration.