AI-Research-SKILLs MCP — Casos de uso, Instalar & Demo en vivo

Por qué usarlo

Características clave

87 skills across 22 categories: model architectures, tokenization, fine-tuning, post-training, distributed training, optimization, inference, safety
Autoresearch orchestrator skill manages full experiment lifecycle
References folder per skill with official docs, GitHub issues, tutorials
Covers modern post-training (TRL, GRPO, SimPO, verl, slime) and inference (vLLM, SGLang, TensorRT-LLM, llama.cpp)

Demo en vivo

Cómo se ve en la práctica

ai-research-skill.replay ▶ listo

0/0

Instalar

Elige tu cliente

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "ai-research-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/Orchestra-Research/AI-Research-SKILLs",
        "~/.claude/skills/AI-Research-SKILLs"
      ],
      "_inferred": true
    }
  }
}

Abre Claude Desktop → Settings → Developer → Edit Config. Reinicia después de guardar.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "ai-research-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/Orchestra-Research/AI-Research-SKILLs",
        "~/.claude/skills/AI-Research-SKILLs"
      ],
      "_inferred": true
    }
  }
}

Cursor usa el mismo esquema mcpServers que Claude Desktop. La configuración del proyecto prevalece sobre la global.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "ai-research-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/Orchestra-Research/AI-Research-SKILLs",
        "~/.claude/skills/AI-Research-SKILLs"
      ],
      "_inferred": true
    }
  }
}

Haz clic en el icono MCP Servers de la barra lateral de Cline y luego en "Edit Configuration".

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "ai-research-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/Orchestra-Research/AI-Research-SKILLs",
        "~/.claude/skills/AI-Research-SKILLs"
      ],
      "_inferred": true
    }
  }
}

Mismo formato que Claude Desktop. Reinicia Windsurf para aplicar.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "ai-research-skill",
      "command": "git",
      "args": [
        "clone",
        "https://github.com/Orchestra-Research/AI-Research-SKILLs",
        "~/.claude/skills/AI-Research-SKILLs"
      ]
    }
  ]
}

Continue usa un array de objetos de servidor en lugar de un mapa.

~/.config/zed/settings.json

{
  "context_servers": {
    "ai-research-skill": {
      "command": {
        "path": "git",
        "args": [
          "clone",
          "https://github.com/Orchestra-Research/AI-Research-SKILLs",
          "~/.claude/skills/AI-Research-SKILLs"
        ]
      }
    }
  }
}

Añádelo a context_servers. Zed recarga en caliente al guardar.

claude mcp add ai-research-skill -- git clone https://github.com/Orchestra-Research/AI-Research-SKILLs ~/.claude/skills/AI-Research-SKILLs

Un solo comando. Verifica con claude mcp list. Quita con claude mcp remove.

Casos de uso

Usos del mundo real: AI-Research-SKILLs

Fine-tune a Llama model with Unsloth + LoRA without Googling

👤 ML engineers running fine-tunes on a single GPU box ⏱ ~20 min intermediate

Cuándo usarlo: You have a dataset and want Claude to produce a runnable training script with correct flags.

Requisitos previos

Unsloth installed — pip install unsloth — the skill will remind you of the nightly build caveats

Flujo

Describe the target run

Use the unsloth skill. Fine-tune Llama-3-8B on my dataset at ~/data/train.jsonl, QLoRA, 3 epochs, output to ./out.✓ Copiado

→ Claude writes a script with correct Unsloth imports and real CLI flags
Ask for the config explanation

Walk me through each hyperparameter and why.✓ Copiado

→ Reasoning grounded in Unsloth's docs, not generic ML clichés

Resultado: A training script that runs on the first try.

Errores comunes

Claude mixes HF Trainer and Unsloth idioms — Insist on 'use the unsloth skill only' — don't pull from Transformers skill

Combinar con: filesystem

Set up a vLLM server with correct tensor-parallelism for your GPUs

👤 Infra engineers serving LLMs in production ⏱ ~30 min advanced

Cuándo usarlo: You have 2-8 GPUs and want Claude to pick the right --tensor-parallel-size and --max-model-len.

Flujo

State the hardware and model

Use the vllm skill. Serve Qwen2.5-72B on 4x H100s. Give me the exact launch command and sanity tests.✓ Copiado

→ Correct TP size and quantization recommendation
Ask for load-test script

Now give me a locust or vllm-benchmark script to verify throughput.✓ Copiado

→ Runnable benchmark using the right endpoint format

Resultado: A vLLM deployment with sanity checks and a benchmark baseline.

Errores comunes

Claude picks a TP size that doesn't divide the attention heads — The vLLM skill references list valid TP sizes per model family — have Claude cite that

Combinar con: aws

Draft an ML paper with correct structure and citation conventions

👤 Researchers writing for NeurIPS/ICML/ICLR ⏱ ~60 min advanced

Cuándo usarlo: You have experiment results and need a paper outline that matches the venue's style.

Flujo

Provide results and venue

Use the ml-paper-writing skill. Target ICLR. Here are my results [paste]. Draft the intro, method, and experiments sections.✓ Copiado

→ Structure follows venue conventions, ablation tables properly framed
Revise for reviewer concerns

What would Reviewer 2 push back on? Add preemptive responses.✓ Copiado

→ Concrete weaknesses, not generic 'add more experiments'

Resultado: A draft that passes initial sanity review and is worth polishing.

Errores comunes

Claude over-claims in the abstract — Explicitly tell it to mirror the restraint of top-tier accepted papers

Combinar con: arxiv

Combinaciones

Combínalo con otros MCPs para multiplicar por 10

ai-research-skill + filesystem

Store training configs, logs, and checkpoints while Claude drives experiments

Save the run config under experiments/<date>-<name>/ and tail the logs.✓ Copiado

ai-research-skill + arxiv

Pull related work while the ml-paper-writing skill drafts sections

Find 5 recent arXiv papers on GRPO and have the paper-writing skill weave them into the related work section.✓ Copiado

Herramientas

Lo que expone este MCP

Herramienta	Entradas	Cuándo llamar
Autoresearch orchestrator	research goal	Complex multi-phase experiments
Fine-tuning (Axolotl, LLaMA-Factory, Unsloth, PEFT)	dataset + base model	SFT / LoRA / QLoRA runs
Post-training (TRL, GRPO, OpenRLHF, SimPO, verl, slime)	reward model or preference data	Alignment, preference optimization
Distributed training (DeepSpeed, FSDP, Megatron, Accelerate)	model + cluster topology	Multi-node or multi-GPU training
Inference (vLLM, SGLang, TensorRT-LLM, llama.cpp)	model + hardware	Serving a model efficiently
Optimization (Flash Attention, bitsandbytes, GPTQ, AWQ, GGUF)	model weights	Fitting a model on smaller hardware

Coste y límites

Lo que cuesta ejecutarlo

Cuota de API: None for the skill itself
Tokens por llamada: Heavy — SKILL.md + references can load 5-10k tokens per sub-skill
Monetario: Free — skills are local files; you pay for compute you run
Consejo: Scope prompts to one sub-skill at a time; the full 87-skill library is too big to load simultaneously.

Seguridad

Permisos, secretos, alcance

Almacenamiento de credenciales: No credentials — skill is prompts + references

Salida de datos: None from the skill itself

Some referenced frameworks have rapid API churn — confirm against the current version if a call fails.

Resolución de problemas

Errores comunes y soluciones

Claude blends two frameworks (e.g. Accelerate + DeepSpeed) incorrectly

Name one skill in the prompt and tell Claude to ignore the other.

References feel outdated

The repo is TeX-heavy and may lag the bleeding edge. For brand-new frameworks, supplement with a fresh fetch of the official docs.

Skill not auto-invoking on a relevant prompt

Mention the framework by name — 87 skills overlap and auto-routing is fuzzy.

Alternativas

AI-Research-SKILLs vs otros

Alternativa	Cuándo usarla	Contrapartida
scientific-agent-skill	You need biology/chemistry/clinical, not ML training infra	Different domain focus
huggingface MCP	You want live HF Hub operations rather than expert prompts	MCP gives you real API actions; this skill teaches the patterns

Más

Recursos

📖 Lee el README oficial en GitHub

🐙 Ver issues abiertas

🔍 Ver todos los 400+ servidores MCP y Skills