# ai-research-skills: a set by orchestra-research

The orchestra-research/ai-research-skills repository on GitHub: 66 cards with a description and an install command. They work in Claude Code, Codex, Cursor — 8 platforms in total.

66 cards · page 2 of 2

- HTML: https://affpapa.org/en/nevibecode/by/orchestra-research/ai-research-skills?page=2
- Instructions for agents: https://affpapa.org/en/nevibecode/agents.md

- [clip](https://affpapa.org/en/nevibecode/skills/clip.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — Use CLIP for zero-shot image classification, image-text matching, and semantic image search without fine-tuning.
- [pytorch-fsdp2](https://affpapa.org/en/nevibecode/skills/pytorch-fsdp2.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — Teaches a coding agent to wire PyTorch FSDP2 into a training loop: init, sharding, mixed precision and distributed checkpoints.
- [awq-quantization](https://affpapa.org/en/nevibecode/skills/awq-quantization.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — awq-quantization — a skill for 4-bit LLM weight quantization with AWQ: faster inference and lower GPU memory use.
- [axolotl](https://affpapa.org/en/nevibecode/skills/axolotl.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — Walks your agent through Axolotl training runs: YAML configs, LoRA and QLoRA, DPO, KTO, ORPO and GRPO methods, dataset formats.
- [huggingface-tokenizers](https://affpapa.org/en/nevibecode/skills/huggingface-tokenizers.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — Provides guides and code patterns for high-speed text tokenization using Hugging Face Tokenizers.
- [sentence-transformers](https://affpapa.org/en/nevibecode/skills/sentence-transformers.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — sentence-transformers guides an agent through local text and image embeddings for RAG, semantic search, clustering, and similarity work.
- [hqq-quantization](https://affpapa.org/en/nevibecode/skills/hqq-quantization.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — hqq-quantization is a skill for Claude Code and Codex that guides agents in quantizing LLM weights to 8/4/3/2/1 bits without calibration data, for serving with vLLM or Transformers.
- [verl-rl-training](https://affpapa.org/en/nevibecode/skills/verl-rl-training.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — verl-rl-training provides step-by-step guidance for reinforcement learning post-training of LLMs using verl with FSDP, Megatron-LM, and Ray.
- [grpo-rl-training](https://affpapa.org/en/nevibecode/skills/grpo-rl-training.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — grpo-rl-training — a skill that walks your agent through GRPO fine-tuning with TRL: reward function templates and training configs.
- [quantizing-models-bitsandbytes](https://affpapa.org/en/nevibecode/skills/quantizing-models-bitsandbytes.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — quantizing-models-bitsandbytes quantizes LLMs to 8-bit or 4-bit via bitsandbytes to cut GPU memory use by 50% to 75%.
- [optimizing-attention-flash](https://affpapa.org/en/nevibecode/skills/optimizing-attention-flash.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — optimizing-attention-flash — a skill that moves transformer attention onto Flash Attention for 2-4x speedup and 10-20x memory reduction.
- [fine-tuning-with-trl](https://affpapa.org/en/nevibecode/skills/fine-tuning-with-trl.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — fine-tuning-with-trl — a skill for Claude Code, Cursor and other platforms adds LLM fine-tuning methods via TRL: SFT for instruction tuning, DPO for preference alignment, PPO and GRPO for reinforcement learning.
- [tensorboard](https://affpapa.org/en/nevibecode/skills/tensorboard.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — This skill gives your agent working recipes for TensorBoard: log training metrics, inspect weight histograms, open the dashboard.
- [qdrant-vector-search](https://affpapa.org/en/nevibecode/skills/qdrant-vector-search.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — This skill guides agents through Qdrant vector search for RAG and semantic search.
- [prompt-guard](https://affpapa.org/en/nevibecode/skills/prompt-guard.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — Integrates Meta's 86M parameter Prompt Guard model to detect jailbreaks and prompt injections in user prompts and RAG contexts before reaching the LLM.
- [unsloth](https://affpapa.org/en/nevibecode/skills/unsloth.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — Skill for Claude Code, Codex, Cursor and other platforms: helps set up fast fine-tuning via Unsloth library.
- [segment-anything-model](https://affpapa.org/en/nevibecode/skills/segment-anything-model.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — A Claude Code skill that covers Meta AI's Segment Anything Model for zero-shot image segmentation.
- [gguf-quantization](https://affpapa.org/en/nevibecode/skills/gguf-quantization.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1100 installs — This skill guides llama.cpp GGUF quantization from 2 to 8 bits.
- [fine-tuning-serving-openpi](https://affpapa.org/en/nevibecode/skills/fine-tuning-serving-openpi.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1000 installs — fine-tuning-serving-openpi is a Claude Code skill with workflows for fine-tuning and serving OpenPI models pi0, pi0-fast, and pi0.5.
- [fine-tuning-openvla-oft](https://affpapa.org/en/nevibecode/skills/fine-tuning-openvla-oft.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1000 installs — fine-tuning-openvla-oft guides fine-tuning and evaluation of OpenVLA-OFT robot action policies.
- [experiment-tracking-swanlab](https://affpapa.org/en/nevibecode/skills/experiment-tracking-swanlab.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 1000 installs — experiment-tracking-swanlab — a skill that shows your agent how to track ML runs in SwanLab: metrics, configs, media logging and a local dashboard.
- [evaluating-cosmos-policy](https://affpapa.org/en/nevibecode/skills/evaluating-cosmos-policy.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 998 installs — evaluating-cosmos-policy — a skill for running NVIDIA Cosmos Policy evaluations on LIBERO and RoboCasa simulators and profiling inference latency.
- [evolving-ai-agents](https://affpapa.org/en/nevibecode/skills/evolving-ai-agents.md) — skill · A-EVO Lab · Claude Code, Codex, Cursor +5 · 972 installs — evolving-ai-agents — a skill for A-Evolve: it runs solve-observe-evolve cycles that rewrite an agent's prompts, skills and memory against benchmark scores.
- [ara-rigor-reviewer](https://affpapa.org/en/nevibecode/skills/ara-rigor-reviewer.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 953 installs — ara-rigor-reviewer reviews an ARA for epistemic rigor before publication or release.
- [ara-compiler](https://affpapa.org/en/nevibecode/skills/ara-compiler.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 935 installs — ara-compiler turns papers, code, experiment logs, and notes into a structured research artifact with claims, evidence, an exploration graph, and code stubs.
- [ara-research-manager](https://affpapa.org/en/nevibecode/skills/ara-research-manager.md) — skill · Orchestra Research · Claude Code, Codex, Cursor +5 · 930 installs — ara-research-manager — a skill that reviews the session after a task is done and writes decisions, experiments and dead ends into the ara/ directory.

## Narrow this list
Parameters: `?platform=<claude-code|codex|cursor|copilot|gemini-cli|windsurf|…>`, `&topic=<…>`, `&sort=installs|stars|new|updated` — values in https://affpapa.org/en/nevibecode/agents.md

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
