orchestra-research/ai-research-skills
99 instruction files for AI agents, collected from orchestra-research/ai-research-skills. Install any of them at an exact version with mdr add orchestra-research/ai-research-skills/<name>.
autoresearch SKILL.md skill
orchestra-research/ai-research-skills · 0-autoresearch-skill/SKILL.md · Orchestrates end-to-end autonomous AI research projects using a two-loop architecture. The inner loop runs rapid…
v1.0.0 · audit A · 13,718 stars
implementing-llms-litgpt SKILL.md skill
orchestra-research/ai-research-skills · 01-model-architecture/litgpt/SKILL.md · Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen…
v1.0.0 · audit A · 13,718 stars
mamba-architecture SKILL.md skill
orchestra-research/ai-research-skills · 01-model-architecture/mamba/SKILL.md · State-space model with O(n) complexity vs Transformers' O(n²). 5× faster inference, million-token sequences, no KV…
v1.0.0 · audit A · 13,718 stars
nanogpt SKILL.md skill
orchestra-research/ai-research-skills · 01-model-architecture/nanogpt/SKILL.md · Educational GPT implementation in ~300 lines. Reproduces GPT-2 (124M) on OpenWebText. Clean, hackable code for learning…
v1.0.0 · audit A · 13,718 stars
rwkv-architecture SKILL.md skill
orchestra-research/ai-research-skills · 01-model-architecture/rwkv/SKILL.md · RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel)…
v1.0.0 · audit A · 13,718 stars
distributed-llm-pretraining-torchtitan SKILL.md skill
orchestra-research/ai-research-skills · 01-model-architecture/torchtitan/SKILL.md · Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when…
v1.0.0 · audit A · 13,718 stars
huggingface-tokenizers SKILL.md skill
orchestra-research/ai-research-skills · 02-tokenization/huggingface-tokenizers/SKILL.md · Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports…
v1.0.0 · audit A · 13,718 stars
sentencepiece SKILL.md skill
orchestra-research/ai-research-skills · 02-tokenization/sentencepiece/SKILL.md · Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k…
v1.0.0 · audit A · 13,718 stars
axolotl SKILL.md skill
orchestra-research/ai-research-skills · 03-fine-tuning/axolotl/SKILL.md · Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO…
v1.0.0 · audit A · 13,718 stars
llama-factory SKILL.md skill
orchestra-research/ai-research-skills · 03-fine-tuning/llama-factory/SKILL.md · Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal…
v1.0.0 · audit A · 13,718 stars
peft-fine-tuning SKILL.md skill
orchestra-research/ai-research-skills · 03-fine-tuning/peft/SKILL.md · Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B)…
v1.0.0 · audit A · 13,718 stars
unsloth SKILL.md skill
orchestra-research/ai-research-skills · 03-fine-tuning/unsloth/SKILL.md · Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization
v1.0.0 · audit A · 13,718 stars
nnsight-remote-interpretability SKILL.md skill
orchestra-research/ai-research-skills · 04-mechanistic-interpretability/nnsight/SKILL.md · Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote…
v1.0.0 · audit A · 13,718 stars
pyvene-interventions SKILL.md skill
orchestra-research/ai-research-skills · 04-mechanistic-interpretability/pyvene/SKILL.md · Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention…
v1.0.0 · audit A · 13,718 stars
sparse-autoencoder-training SKILL.md skill
orchestra-research/ai-research-skills · 04-mechanistic-interpretability/saelens/SKILL.md · Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network…
v1.0.0 · audit A · 13,718 stars
transformer-lens-interpretability SKILL.md skill
orchestra-research/ai-research-skills · 04-mechanistic-interpretability/transformer-lens/SKILL.md · Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer…
v1.0.0 · audit A · 13,718 stars
nemo-curator SKILL.md skill
orchestra-research/ai-research-skills · 05-data-processing/nemo-curator/SKILL.md · GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16×…
v1.0.0 · audit A · 13,718 stars
ray-data SKILL.md skill
orchestra-research/ai-research-skills · 05-data-processing/ray-data/SKILL.md · Scalable data processing for ML workloads. Streaming execution across CPU/GPU, supports Parquet/CSV/JSON/images…
v1.0.0 · audit A · 13,718 stars
grpo-rl-training SKILL.md skill
orchestra-research/ai-research-skills · 06-post-training/grpo-rl-training/SKILL.md · Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training
v1.0.0 · audit A · 13,718 stars
miles-rl-training SKILL.md skill
orchestra-research/ai-research-skills · 06-post-training/miles/SKILL.md · Provides guidance for enterprise-grade RL training using miles, a production-ready fork of slime. Use when training…
v1.0.0 · audit A · 13,718 stars
openrlhf-training SKILL.md skill
orchestra-research/ai-research-skills · 06-post-training/openrlhf/SKILL.md · High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models…
v1.0.0 · audit B · 13,718 stars
simpo-training SKILL.md skill
orchestra-research/ai-research-skills · 06-post-training/simpo/SKILL.md · Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4…
v1.0.0 · audit A · 13,718 stars
slime-rl-training SKILL.md skill
orchestra-research/ai-research-skills · 06-post-training/slime/SKILL.md · Provides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM models…
v1.0.0 · audit A · 13,718 stars
torchforge-rl-training SKILL.md skill
orchestra-research/ai-research-skills · 06-post-training/torchforge/SKILL.md · Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use…
v1.0.0 · audit A · 13,718 stars
fine-tuning-with-trl SKILL.md skill
orchestra-research/ai-research-skills · 06-post-training/trl-fine-tuning/SKILL.md · Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment…
v1.0.0 · audit A · 13,718 stars
verl-rl-training SKILL.md skill
orchestra-research/ai-research-skills · 06-post-training/verl/SKILL.md · Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing…
v1.0.0 · audit A · 13,718 stars
constitutional-ai SKILL.md skill
orchestra-research/ai-research-skills · 07-safety-alignment/constitutional-ai/SKILL.md · Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with…
v1.0.0 · audit A · 13,718 stars
llamaguard SKILL.md skill
orchestra-research/ai-research-skills · 07-safety-alignment/llamaguard/SKILL.md · Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual…
v1.0.0 · audit A · 13,718 stars
nemo-guardrails SKILL.md skill
orchestra-research/ai-research-skills · 07-safety-alignment/nemo-guardrails/SKILL.md · NVIDIA's runtime safety framework for LLM applications. Features jailbreak detection, input/output validation…
v1.0.0 · audit B · 13,718 stars
prompt-guard SKILL.md skill
orchestra-research/ai-research-skills · 07-safety-alignment/prompt-guard/SKILL.md · Meta's 86M prompt injection and jailbreak detector. Filters malicious prompts and third-party data for LLM apps. 99%+…
v1.0.0 · audit B · 13,718 stars
huggingface-accelerate SKILL.md skill
orchestra-research/ai-research-skills · 08-distributed-training/accelerate/SKILL.md · Simplest distributed training API. 4 lines to add distributed support to any PyTorch script. Unified API for…
v1.0.0 · audit A · 13,718 stars
deepspeed SKILL.md skill
orchestra-research/ai-research-skills · 08-distributed-training/deepspeed/SKILL.md · Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism…
v1.0.0 · audit A · 13,718 stars
training-llms-megatron SKILL.md skill
orchestra-research/ai-research-skills · 08-distributed-training/megatron-core/SKILL.md · Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use…
v1.0.0 · audit A · 13,718 stars
pytorch-fsdp2 SKILL.md skill
orchestra-research/ai-research-skills · 08-distributed-training/pytorch-fsdp2/SKILL.md · Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and…
v1.0.0 · audit A · 13,718 stars
pytorch-lightning SKILL.md skill
orchestra-research/ai-research-skills · 08-distributed-training/pytorch-lightning/SKILL.md · High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system…
v1.0.0 · audit A · 13,718 stars
ray-train SKILL.md skill
orchestra-research/ai-research-skills · 08-distributed-training/ray-train/SKILL.md · Distributed training orchestration across clusters. Scales PyTorch/TensorFlow/HuggingFace from laptop to 1000s of…
v1.0.0 · audit A · 13,718 stars
lambda-labs-gpu-cloud SKILL.md skill
orchestra-research/ai-research-skills · 09-infrastructure/lambda-labs/SKILL.md · Reserved and on-demand GPU cloud instances for ML training and inference. Use when you need dedicated GPU instances…
v1.0.0 · audit A · 13,718 stars
modal-serverless-gpu SKILL.md skill
orchestra-research/ai-research-skills · 09-infrastructure/modal/SKILL.md · Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure…
v1.0.0 · audit A · 13,718 stars
skypilot-multi-cloud-orchestration SKILL.md skill
orchestra-research/ai-research-skills · 09-infrastructure/skypilot/SKILL.md · Multi-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch…
v1.0.0 · audit A · 13,718 stars
awq-quantization SKILL.md skill
orchestra-research/ai-research-skills · 10-optimization/awq/SKILL.md · Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when…
v1.0.0 · audit A · 13,718 stars
quantizing-models-bitsandbytes SKILL.md skill
orchestra-research/ai-research-skills · 10-optimization/bitsandbytes/SKILL.md · Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is…
v1.0.0 · audit A · 13,718 stars
optimizing-attention-flash SKILL.md skill
orchestra-research/ai-research-skills · 10-optimization/flash-attention/SKILL.md · Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Use when…
v1.0.0 · audit A · 13,718 stars
gguf-quantization SKILL.md skill
orchestra-research/ai-research-skills · 10-optimization/gguf/SKILL.md · GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware…
v1.0.0 · audit A · 13,718 stars
gptq SKILL.md skill
orchestra-research/ai-research-skills · 10-optimization/gptq/SKILL.md · Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on…
v1.0.0 · audit A · 13,718 stars
hqq-quantization SKILL.md skill
orchestra-research/ai-research-skills · 10-optimization/hqq/SKILL.md · Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision…
v1.0.0 · audit A · 13,718 stars
ml-training-recipes SKILL.md skill
orchestra-research/ai-research-skills · 10-optimization/ml-training-recipes/SKILL.md · Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug…
v1.0.0 · audit A · 13,718 stars
evaluating-code-models SKILL.md skill
orchestra-research/ai-research-skills · 11-evaluation/bigcode-evaluation-harness/SKILL.md · Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when…
v1.0.0 · audit A · 13,718 stars
evaluating-llms-harness SKILL.md skill
orchestra-research/ai-research-skills · 11-evaluation/lm-evaluation-harness/SKILL.md · Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking…
v1.0.0 · audit A · 13,718 stars
nemo-evaluator-sdk SKILL.md skill
orchestra-research/ai-research-skills · 11-evaluation/nemo-evaluator/SKILL.md · Evaluates LLMs across 100+ benchmarks from 18+ harnesses (MMLU, HumanEval, GSM8K, safety, VLM) with multi-backend…
v1.0.0 · audit A · 13,718 stars
llama-cpp SKILL.md skill
orchestra-research/ai-research-skills · 12-inference-serving/llama-cpp/SKILL.md · Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3…
v1.0.0 · audit A · 13,718 stars
sglang SKILL.md skill
orchestra-research/ai-research-skills · 12-inference-serving/sglang/SKILL.md · Fast structured generation and serving for LLMs with RadixAttention prefix caching. Use for JSON/regex outputs…
v1.0.0 · audit A · 13,718 stars
tensorrt-llm SKILL.md skill
orchestra-research/ai-research-skills · 12-inference-serving/tensorrt-llm/SKILL.md · Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment…
v1.0.0 · audit A · 13,718 stars
serving-llms-vllm SKILL.md skill
orchestra-research/ai-research-skills · 12-inference-serving/vllm/SKILL.md · Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM…
v1.0.0 · audit A · 13,718 stars
mlflow SKILL.md skill
orchestra-research/ai-research-skills · 13-mlops/mlflow/SKILL.md · Track ML experiments, manage model registry with versioning, deploy models to production, and reproduce experiments…
v1.0.0 · audit B · 13,718 stars
experiment-tracking-swanlab SKILL.md skill
orchestra-research/ai-research-skills · 13-mlops/swanlab/SKILL.md · Provides guidance for experiment tracking with SwanLab. Use when you need open-source run tracking, local or…
v1.0.0 · audit A · 13,718 stars
tensorboard SKILL.md skill
orchestra-research/ai-research-skills · 13-mlops/tensorboard/SKILL.md · Visualize training metrics, debug models with histograms, compare experiments, visualize model graphs, and profile…
v1.0.0 · audit A · 13,718 stars
weights-and-biases SKILL.md skill
orchestra-research/ai-research-skills · 13-mlops/weights-and-biases/SKILL.md · Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and…
v1.0.0 · audit A · 13,718 stars
evolving-ai-agents SKILL.md skill
orchestra-research/ai-research-skills · 14-agents/a-evolve/SKILL.md · Provides guidance for automatically evolving and optimizing AI agents across any domain using LLM-driven evolution…
v1.0.0 · audit A · 13,718 stars
autogpt-agents SKILL.md skill
orchestra-research/ai-research-skills · 14-agents/autogpt/SKILL.md · Autonomous AI agent platform for building and deploying continuous agents. Use when creating visual workflow agents…
v1.0.0 · audit A · 13,718 stars
crewai-multi-agent SKILL.md skill
orchestra-research/ai-research-skills · 14-agents/crewai/SKILL.md · Multi-agent orchestration framework for autonomous AI collaboration. Use when building teams of specialized agents…
v1.0.0 · audit A · 13,718 stars
langchain SKILL.md skill
orchestra-research/ai-research-skills · 14-agents/langchain/SKILL.md · Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI…
v1.0.0 · audit A · 13,718 stars
llamaindex SKILL.md skill
orchestra-research/ai-research-skills · 14-agents/llamaindex/SKILL.md · Data framework for building LLM applications with RAG. Specializes in document ingestion (300+ connectors), indexing…
v1.0.0 · audit A · 13,718 stars
chroma SKILL.md skill
orchestra-research/ai-research-skills · 15-rag/chroma/SKILL.md · Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search…
v1.0.0 · audit A · 13,718 stars
faiss SKILL.md skill
orchestra-research/ai-research-skills · 15-rag/faiss/SKILL.md · Facebook's library for efficient similarity search and clustering of dense vectors. Supports billions of vectors, GPU…
v1.0.0 · audit A · 13,718 stars
pinecone SKILL.md skill
orchestra-research/ai-research-skills · 15-rag/pinecone/SKILL.md · Managed vector database for production AI applications. Fully managed, auto-scaling, with hybrid search (dense +…
v1.0.0 · audit A · 13,718 stars
qdrant-vector-search SKILL.md skill
orchestra-research/ai-research-skills · 15-rag/qdrant/SKILL.md · High-performance vector similarity search engine for RAG and semantic search. Use when building production RAG systems…
v1.0.0 · audit A · 13,718 stars
sentence-transformers SKILL.md skill
orchestra-research/ai-research-skills · 15-rag/sentence-transformers/SKILL.md · Framework for state-of-the-art sentence, text, and image embeddings. Provides 5000+ pre-trained models for semantic…
v1.0.0 · audit A · 13,718 stars
dspy SKILL.md skill
orchestra-research/ai-research-skills · 16-prompt-engineering/dspy/SKILL.md · Build complex AI systems with declarative programming, optimize prompts automatically, create modular RAG systems and…
v1.0.0 · audit A · 13,718 stars
guidance SKILL.md skill
orchestra-research/ai-research-skills · 16-prompt-engineering/guidance/SKILL.md · Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and…
v1.0.0 · audit A · 13,718 stars
instructor SKILL.md skill
orchestra-research/ai-research-skills · 16-prompt-engineering/instructor/SKILL.md · Extract structured data from LLM responses with Pydantic validation, retry failed extractions automatically, parse…
v1.0.0 · audit A · 13,718 stars
outlines SKILL.md skill
orchestra-research/ai-research-skills · 16-prompt-engineering/outlines/SKILL.md · Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local…
v1.0.0 · audit A · 13,718 stars
langsmith-observability SKILL.md skill
orchestra-research/ai-research-skills · 17-observability/langsmith/SKILL.md · LLM observability platform for tracing, evaluation, and monitoring. Use when debugging LLM applications, evaluating…
v1.0.0 · audit A · 13,718 stars
phoenix-observability SKILL.md skill
orchestra-research/ai-research-skills · 17-observability/phoenix/SKILL.md · Open-source AI observability platform for LLM tracing, evaluation, and monitoring. Use when debugging LLM applications…
v1.0.0 · audit A · 13,718 stars
audiocraft-audio-generation SKILL.md skill
orchestra-research/ai-research-skills · 18-multimodal/audiocraft/SKILL.md · PyTorch library for audio generation including text-to-music (MusicGen) and text-to-sound (AudioGen). Use when you need…
v1.0.0 · audit A · 13,718 stars
blip-2-vision-language SKILL.md skill
orchestra-research/ai-research-skills · 18-multimodal/blip-2/SKILL.md · Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning…
v1.0.0 · audit A · 13,718 stars
clip SKILL.md skill
orchestra-research/ai-research-skills · 18-multimodal/clip/SKILL.md · OpenAI's model connecting vision and language. Enables zero-shot image classification, image-text matching, and…
v1.0.0 · audit A · 13,718 stars
evaluating-cosmos-policy SKILL.md skill
orchestra-research/ai-research-skills · 18-multimodal/cosmos-policy/SKILL.md · Evaluates NVIDIA Cosmos Policy on LIBERO and RoboCasa simulation environments. Use when setting up cosmos-policy for…
v1.0.0 · audit A · 13,718 stars
llava SKILL.md skill
orchestra-research/ai-research-skills · 18-multimodal/llava/SKILL.md · Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP…
v1.0.0 · audit A · 13,718 stars
fine-tuning-serving-openpi SKILL.md skill
orchestra-research/ai-research-skills · 18-multimodal/openpi/SKILL.md · Fine-tune and serve Physical Intelligence OpenPI models (pi0, pi0-fast, pi0.5) using JAX or PyTorch backends for robot…
v1.0.0 · audit A · 13,718 stars
fine-tuning-openvla-oft SKILL.md skill
orchestra-research/ai-research-skills · 18-multimodal/openvla-oft/SKILL.md · Fine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies for robot action generation with continuous action…
v1.0.0 · audit A · 13,718 stars
segment-anything-model SKILL.md skill
orchestra-research/ai-research-skills · 18-multimodal/segment-anything/SKILL.md · Foundation model for image segmentation with zero-shot transfer. Use when you need to segment any object in images…
v1.0.0 · audit A · 13,718 stars
stable-diffusion-image-generation SKILL.md skill
orchestra-research/ai-research-skills · 18-multimodal/stable-diffusion/SKILL.md · State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating…
v1.0.0 · audit A · 13,718 stars
whisper SKILL.md skill
orchestra-research/ai-research-skills · 18-multimodal/whisper/SKILL.md · OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and…
v1.0.0 · audit A · 13,718 stars
knowledge-distillation SKILL.md skill
orchestra-research/ai-research-skills · 19-emerging-techniques/knowledge-distillation/SKILL.md · Compress large language models using knowledge distillation from teacher to student models. Use when deploying smaller…
v1.0.0 · audit A · 13,718 stars
long-context SKILL.md skill
orchestra-research/ai-research-skills · 19-emerging-techniques/long-context/SKILL.md · Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques. Use when…
v1.0.0 · audit A · 13,718 stars
model-merging SKILL.md skill
orchestra-research/ai-research-skills · 19-emerging-techniques/model-merging/SKILL.md · Merge multiple fine-tuned models using mergekit to combine capabilities without retraining. Use when creating…
v1.0.0 · audit A · 13,718 stars
model-pruning SKILL.md skill
orchestra-research/ai-research-skills · 19-emerging-techniques/model-pruning/SKILL.md · Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT. Use when compressing models…
v1.0.0 · audit A · 13,718 stars
moe-training SKILL.md skill
orchestra-research/ai-research-skills · 19-emerging-techniques/moe-training/SKILL.md · Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited…
v1.0.0 · audit A · 13,718 stars
speculative-decoding SKILL.md skill
orchestra-research/ai-research-skills · 19-emerging-techniques/speculative-decoding/SKILL.md · Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when…
v1.0.0 · audit A · 13,718 stars
academic-plotting SKILL.md skill
orchestra-research/ai-research-skills · 20-ml-paper-writing/academic-plotting/SKILL.md · Generates publication-quality figures for ML papers from research context. Given a paper section or description…
v1.0.0 · audit A · 13,718 stars
ml-paper-writing SKILL.md skill
orchestra-research/ai-research-skills · 20-ml-paper-writing/ml-paper-writing/SKILL.md · Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research…
v1.2.0 · audit A · 13,718 stars
presenting-conference-talks SKILL.md skill
orchestra-research/ai-research-skills · 20-ml-paper-writing/presenting-conference-talks/SKILL.md · Generates conference presentation slides (Beamer LaTeX PDF and editable PPTX) from a compiled paper with speaker notes…
v1.0.0 · audit A · 13,718 stars
systems-paper-writing SKILL.md skill
orchestra-research/ai-research-skills · 20-ml-paper-writing/systems-paper-writing/SKILL.md · Comprehensive guide for writing systems papers targeting OSDI, SOSP, ASPLOS, NSDI, and EuroSys. Provides…
v1.1.0 · audit A · 13,718 stars
brainstorming-research-ideas SKILL.md skill
orchestra-research/ai-research-skills · 21-research-ideation/brainstorming-research-ideas/SKILL.md · Guides researchers through structured ideation frameworks to discover high-impact research directions. Use when…
v1.0.0 · audit A · 13,718 stars
creative-thinking-for-research SKILL.md skill
orchestra-research/ai-research-skills · 21-research-ideation/creative-thinking-for-research/SKILL.md · Applies cognitive science frameworks for creative thinking to CS and AI research ideation. Use when seeking genuinely…
v1.0.0 · audit A · 13,718 stars
ara-compiler SKILL.md skill
orchestra-research/ai-research-skills · 22-agent-native-research-artifact/compiler/SKILL.md · Compiles any research input — PDF papers, GitHub repositories, experiment logs, code directories, or raw notes — into a…
v1.0.0 · audit A · 13,718 stars
ara-research-manager SKILL.md skill
orchestra-research/ai-research-skills · 22-agent-native-research-artifact/research-manager/SKILL.md · Records research provenance as a post-task epilogue, scanning conversation history at the end of a coding or research…
v1.0.0 · audit A · 13,718 stars
ara-rigor-reviewer SKILL.md skill
orchestra-research/ai-research-skills · 22-agent-native-research-artifact/rigor-reviewer/SKILL.md · Performs ARA Seal Level 2 semantic epistemic review on Agent-Native Research Artifacts, scoring six dimensions…
v3.0.0 · audit A · 13,718 stars
git:20260615.f8afc97 · audit A · 13,718 stars
Source
github.com/orchestra-research/ai-research-skills · 13,718 stars · license MIT · pushed 2026-10-11