# Orchestra-Research/AI-Research-SKILLs

Comprehensive open-source library of AI research and engineering skills for any AI model. Package the skills and your claude code/codex/gemini agent will be an AI research agent with full horsepower. Maintained by Orchestra Research.

Repository: https://github.com/Orchestra-Research/AI-Research-SKILLs
Canonical: https://ross.abutalabs.com/products/ai-research-skills
Homepage: http://orchestra-research.com
Language: TeX
License: MIT
License Family: permissive
Topics: ai, ai-research, claude, claude-code, claude-skills, codex, gemini, gpt-5, grpo, huggingface, machine-leanring, megatron, skills, vllm
Last push: 2026-06-16T01:36:46+00:00

## Health v2 (maintenance only)
Score: 74/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 87, release rhythm 89, longevity 21
- inputs: {"age_days": 303, "days_push": 79, "days_rel": 79, "gap_med": 3.5, "n_releases_24m": 9}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12073, forks 877 (observed 2026-08-28T04:10:51.660128+00:00)

## What it is
An open-source library of ~98 packaged 'skills' (knowledge modules) that turn AI coding agents like Claude Code, Codex, or Gemini into autonomous AI research agents, covering the full lifecycle from ideation and literature review to training, evaluation, and paper writing. It is maintained by Orchestra Research and distributed via npm for use with popular agent harnesses.

## Use cases
- turn claude code into an ai research agent
- automate ml experiments with an ai agent
- have an agent write and run fine-tuning experiments
- automated literature review and paper writing
- skills for vllm and megatron training workflows
- run grpo reinforcement learning fine-tuning with an agent
- package research knowledge for codex or gemini agents

## When to choose
- you use Claude Code, Codex, or Gemini CLI and want agent-driven ML research workflows
- you need structured, reusable knowledge for training, evaluation, RAG, or paper writing inside an agent
- you want to accelerate the idea-to-paper research pipeline with automation

## When to avoid
- you need a turnkey research platform with managed GPUs rather than a skills library (consider Orchestra's paid platform instead)
- your work is outside AI/ML research domains
- you want a traditional library you call from application code rather than agent-consumable skill packages

## Facets
- artifact type: library
- maturity: active
- function: agent-framework, prompt-engineering, llm-training, rag, machine-learning, documentation, developer-tools
- domain: artificial-intelligence, machine-learning, large-language-models, deep-learning, developer-tools, tutorials
- platform: python, cross-platform, cli
- tags: claude-skills, claude-code, codex, gemini, vllm, megatron, huggingface, grpo, research-automation, agent-skills, paper-writing, fine-tuning, ai-agents

## Member repositories
- Orchestra-Research/AI-Research-SKILLs (main) score 74

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:51.660128+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:14:57.653645+00:00, confidence not recorded.
  - readme: https://github.com/Orchestra-Research/AI-Research-SKILLs (fetched 2026-08-28T04:10:51.660128+00:00, sha d2fdda2a45dc)
  - homepage: http://orchestra-research.com (fetched 2026-08-29T08:12:26.087563+00:00, sha 410752cb7994)
  - site_page: https://www.orchestra-research.com/pricing-public (fetched 2026-08-29T08:12:26.096697+00:00, sha ac0df02b89f6)
- Data as of 2026-08-30T08:39:29.467469+00:00.