Orchestra-ResearchOfficial·85 Agent Skills Included

AI-Research-SKILLs

98 skills for autonomous AI research from idea to paper

Guides AI agents through the full research lifecycle: literature survey, ideation, data processing, model training, evaluation, deployment, and paper writing. Covers 98 production-ready skills across 23 categories including fine-tuning, RLHF, inference serving, RAG, and LaTeX paper templates. Eliminates manual framework debugging with expert workflows, real code examples, and troubleshooting guides for each tool.

11,537|842

↗

npx skills add Orchestra-Research/AI-Research-SKILLs --all -g -y

Available:

Explains the repository's 23-category skill structure and quality standards so the AI agent can navigate, use, and extend the research skills correctly.

All Skills in This Repository (85)

Pure Emerald Level Indicators

📦 In Repo

Orchestra-Research

ml-paper-writing

Draft machine learning conference papers with LaTeX templates and API-verified citations.

serving-llms-vllm

Serve LLMs with vLLM using PagedAttention and continuous batching.

tensorrt-llm

Optimize LLM inference on NVIDIA GPUs using TensorRT-LLM with quantization and batching.

sglang

Serve LLMs with RadixAttention prefix caching and structured JSON outputs.

llama-cpp

Execute LLM inference on CPUs, Apple Silicon, and non-NVIDIA GPUs using llama.cpp.

lambda-labs-gpu-cloud

Provision on-demand and reserved NVIDIA GPU cloud instances for ML workloads.

skypilot-multi-cloud-orchestration

Orchestrate machine learning workloads across AWS, GCP, Azure, and Kubernetes.

modal-serverless-gpu

Run machine learning workloads on serverless GPU compute with auto-scaling APIs.

nemo-curator

Prepare LLM training data with GPU-accelerated NeMo Curator.

ray-data

Process large datasets with distributed transformations across CPU and GPU resources.

dspy

Compile high-level Python specifications into optimized LLM calls.

guidance

Control LLM generation with Pythonic syntax, regex, and context-free grammars.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install AI-Research-SKILLs?▼

Run `npx skills add Orchestra-Research/AI-Research-SKILLs --all -g -y` in your terminal to install all 98 skills globally.

What can AI-Research-SKILLs do?▼

It lets your AI agent autonomously run the full research lifecycle: literature search, hypothesis generation, experiments, model training, evaluation, and writing conference-ready papers.

Which frameworks are covered?▼

Skills cover vLLM, TensorRT-LLM, DeepSpeed, FSDP, TRL, Axolotl, LangChain, DSPy, Chroma, Qdrant, Stable Diffusion, Whisper, and dozens more across training, inference, RAG, and multimodal work.

Does it work with Claude Code and other agents?▼

Yes. All skills follow the standard SKILL.md format and work with Claude Code, OpenClaw, Cursor, and other compatible agents, including an autoresearch orchestrator for continuous autonomous operation.

Can it help write ML conference papers?▼

Yes. It includes paper-writing skills with LaTeX templates for NeurIPS, ICML, ICLR, OSDI, and SOSP, plus citation verification workflows and academic figure generation.