VectorSpaceLab/AREX-Skill

dynamic-resources

Example skill loaded from resources_discover

원문 언어: 영어

create-paper-skills

Creates reusable Agent Skills from an AI research paper using the Paper2Skills Distiller workflow. Use in Creator mode when the user provides a paper PDF, paper URL, arXiv id, paper title, or paper/repo pair, or asks to convert a scientific paper into skills…

원문 언어: 영어

create-repo-skill

Creates a repo-specific operating Agent Skill for DisCo Researcher from a local repository by inspecting source files and an installed or auto-prepared, backend-aware Python package environment. Use when the user asks to create a skill for a repo, generate…

원문 언어: 영어

extend-repo-skill

Extends an existing repository-specific Agent Skill with new capabilities, deeper coverage, troubleshooting, scripts, and usability tests. Use when the user asks to expand, improve, deepen, or add coverage to an already implemented skill instead of creating a…

원문 언어: 영어

import-repo-skills-to-agent

Use this skill when the user asks to export DisCo's managed repository skills and repo-skills-router into another agent tool such as Codex under ~/.agents/skills, Claude Code under ~/.claude/skills, or a project-local agent directory. Handles canonical source…

원문 언어: 영어

refresh-repo-skill

Refreshes an existing repository-specific Agent Skill after the source repository changed. Use when the user says repo code, APIs, docs, examples, configs, dependencies, or behavior changed and an old skill may now be stale, outdated, inconsistent with…

원문 언어: 영어

repo-skills-router

Routes substantive ML, AI, data, scientific-computing, and software-engineering requests to the smallest useful set of managed repository skills. Invoke proactively when a request names or implies a package, framework, model family, dataset, modality,…

원문 언어: 영어

verify-repo-skill

Verifies a generated or refreshed repo-specific Agent Skill by creating assertion-backed usability test cases, running content-level self-refine, checking backend-classified native repo examples/tests against the prepared CPU/GPU environment plan, enforcing…

원문 언어: 영어

3d-resnets-pytorch

Routes 3D ResNets PyTorch video action-recognition workflows across training, inference, and data preparation.

원문 언어: 영어

data-preparation

Prepare raw videos and annotations for the 3D-ResNets-PyTorch dataset loaders.

원문 언어: 영어

training-and-inference

Routes training, fine-tuning, validation, checkpoint, and inference workflows for 3D ResNets PyTorch.

원문 언어: 영어

3ddfa

Guide 3DDFA Python inference, geometry rendering, training/evaluation, and optional C++ ONNX workflows for 3D dense face alignment.

원문 언어: 영어

cpp-onnx-port

Operate the optional 3DDFA C++ OpenCV DNN port, checkpoint-to-ONNX export, weight placement, build/run expectations, and C++ troubleshooting.

원문 언어: 영어

geometry-rendering

Reconstruct 3DMM vertices, serialize mesh outputs, and manage 3DDFA rendering helpers.

원문 언어: 영어

python-inference

Operate 3DDFA Python image and video inference for landmarks, dense vertices, meshes, pose boxes, depth, PNCC, and PAF outputs.

원문 언어: 영어

training-evaluation

Routes 3DDFA training recipes, loss selection, checkpoint resume, dataset layout, and benchmark evaluation.

원문 언어: 영어

3ddfa-v2

Routes 3DDFA_V2 face-alignment setup, still-image demos, video tracking, and ONNX benchmarking workflows.

원문 언어: 영어

onnx-and-benchmarking

Use 3DDFA_V2 ONNX Runtime acceleration and CPU latency or speed benchmark workflows.

원문 언어: 영어

setup-and-assets

Prepare 3DDFA_V2 runtime assets and native extension builds before demos or benchmarks.

원문 언어: 영어

still-image-demo

Run 3DDFA_V2 still-image alignment, rendering, pose, texture, and mesh export workflows.

원문 언어: 영어

video-and-tracking

Run 3DDFA_V2 video, smoothing, and manual webcam tracking workflows.

원문 언어: 영어

ab3dmot

Operate AB3DMOT 3D multi-object tracking workflows for KITTI and nuScenes data, tracking, evaluation, and visualization.

원문 언어: 영어

data-conversion

Routes KITTI and nuScenes data-layout, detection-conversion, and schema-validation work for AB3DMOT inputs.

원문 언어: 영어

evaluation-visualization

Evaluate, threshold, combine, and visualize AB3DMOT KITTI and nuScenes tracking results.

원문 언어: 영어

tracking-pipeline

Run AB3DMOT tracking safely and use the core AB3DMOT tracker APIs for KITTI and nuScenes 3D MOT workflows.

원문 언어: 영어

accelerate

Use Hugging Face Accelerate for PyTorch training-loop migration, distributed launch/configuration, DeepSpeed/FSDP/TPU backend setup, big-model inference/offload, checkpointing, tracking, and troubleshooting.

원문 언어: 영어

big-model-inference

Use Accelerate big-model inference utilities for meta initialization, device-map planning, checkpoint dispatch, CPU/disk offload, hooks, pipeline inference, and memory sizing without triggering downloads or heavyweight runs.

원문 언어: 영어

checkpointing-and-tracking

Save and resume Accelerate training state, register checkpoint hooks and custom state, log safely across processes, use experiment trackers, profile runs, and clean up memory.

원문 언어: 영어

configuration-and-cli

Use this sub-skill when working with Hugging Face Accelerate configuration files and CLI commands, including accelerate config/default/update/env/launch/test/estimate-memory/merge-weights/to-fsdp2, launch command construction, multi-node and SLURM planning,…

원문 언어: 영어

distributed-training-backends

Select, configure, and diagnose Accelerate distributed training backends including DeepSpeed, FSDP/FSDP2, Megatron-LM, torch native parallelism, TPU/XLA, FP8, quantization, compilation, Local SGD, and DDP communication hooks.

원문 언어: 영어

training-loop-integration

Migrate raw PyTorch training and evaluation loops to Hugging Face Accelerate using Accelerator, prepare(), backward(), gradient accumulation, dataloader behavior, gather/reduce, mixed precision, DDP kwargs, local SGD, communication hooks, and basic…

원문 언어: 영어

acme

Route Acme reinforcement-learning framework tasks across core loops, replay/data, JAX agents, and TensorFlow/Sonnet agents.

원문 언어: 영어

core-workflows

Build and debug Acme core dm_env loops, specs, wrappers, logging, counting, observers, and simple custom Actor/Learner components.

원문 언어: 영어

jax-agents

Select, configure, and adapt Acme JAX agents and JAX experiment workflows.

원문 언어: 영어

replay-and-data

Use Acme adders, Reverb replay tables and datasets, offline data iterators, image augmentation, and replay shape troubleshooting.

원문 언어: 영어

tf-agents

Select, configure, debug, and adapt Acme TensorFlow/Sonnet agents, networks, savers, Launchpad examples, and TF losses.

원문 언어: 영어

act-plus-plus

Routes ACT++, ACT, Diffusion Policy, VINN, and MuJoCo simulation workflows for bimanual ALOHA episode data and imitation-learning tasks.

원문 언어: 영어

policy-training

Routes ACT, CNNMLP, Diffusion Policy, and latent-model training or evaluation workflows for ACT++ checkpoints and datasets.

원문 언어: 영어

simulation-data

Routes simulated ALOHA episode generation, replay, visualization, mirroring, compression, and truncation workflows for ACT++ HDF5 data.

원문 언어: 영어

vinn-offline

Routes VINN feature caching and non-interactive k-selection workflows for ACT++ BYOL/ResNet episode features.

원문 언어: 영어