RadarTopicsBuildersWeeklyReads
Open Source Radar
Orchestra-Research/

AI-Research-SKILLs

GitHubWebsite

A large, MIT-licensed TeX-based repository hosting an extensive, 98-skill AI research library for autonomous agent workflows, with frequent releases and cross-agent compatibility.

11kstars
833forks
17issues
MITlicense
2025since
Star historydaily snapshots by VibeCrowd

Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).

Alternatives & relatedmatched by topic overlap
Reviewgenerated from repository data · Aug 5, 2026

What it is

The project is an open-source library of AI research and engineering skills intended to enable AI agents to autonomously conduct AI research from idea to paper. It describes a collection of skills across 23 categories and an orchestration layer called autoresearch.

How it works

The library provides 98 skills organized into 23 categories, plus an autoresearch orchestration layer. It includes an installer that detects installed agents and installs skills to a user directory, with symlinks to agent-specific skill sets. It supports both interactive and direct CLI usage, and a Claude Code Marketplace pathway for installation by category.

Getting started

Quick Install (Recommended)

npx @orchestra-research/ai-research-skills

For AI agents — follow the welcome documentation:

Read https://www.orchestra-research.com/ai-research-skills/welcome.md? and follow the instructions to install and use AI Research Skills.

CLI Commands

# Interactive installer (recommended)
npx @orchestra-research/ai-research-skills

# Direct commands
npx @orchestra-research/ai-research-skills list      # View installed skills
npx @orchestra-research/ai-research-skills update    # Update installed skills

Claude Code Marketplace (Alternative)

/plugin marketplace add orchestra-research/AI-research-SKILLs
/plugin install fine-tuning@ai-research-skills
/plugin install post-training@ai-research-skills
/plugin install inference-serving@ai-research-skills
/plugin install distributed-training@ai-research-skills
/plugin install optimization@ai-research-skills

The installer places skills under ~/.orchestra/skills/ and supports symlinks to agent directories.

Recent releases

Latest releases include:

  • v1.7.2 (2026-06-16): Qoder agent support, installer auto-detects Qoder, installs skills to ~/.qoder/skills/.
  • v1.7.1 (2026-06-15): Inventory hardened; 98-skill content milestone shipped in v1.6.0.
  • v1.4.0 (2026-03-16): Autoresearch feature introduced for end-to-end autonomous research orchestration.
  • v1.3.6 (2026-02-08): Local project installation flag for per-project skills.
  • v1.2.0 (2026-02-06): OpenClaw support and shared .agents/ directory for skills.

Traction

Stars: 11391 Forks: 833 Open issues: 17

Behind the repo

Not provided in the README excerpt.

Caveats

License: MIT Created: 2025-11-03 Last push: 2026-06-16 Language: TeX Topics include ai, ai-research, claude, claude-code, claude-skills, codex, gemini, gpt-5, grpo, huggingface, machine-learning, megatron, skills, vllm

Agent skills inside · 98
All skills →
grpo-rl-trainingExpert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model trainingWorkflow & Productivityscriptsml-paper-writingWrite publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. For systems venues (OSDI, NSDI, ASPLOS, SOSP), use systems-paper-writing instead.Documentationsystems-paper-writingComprehensive guide for writing systems papers targeting OSDI, SOSP, ASPLOS, NSDI, and EuroSys. Provides paragraph-level structural blueprints, writing patterns, venue-specific checklists, reviewer guidelines, LaTeX templates, and conference deadlines. Use this skill for all systems conference paper writing.Documentationpytorch-fsdp2Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you need DTensor-based sharding with DeviceMesh.Data & AnalyticsdeepspeedExpert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attentionOtherevolving-ai-agentsProvides guidance for automatically evolving and optimizing AI agents across any domain using LLM-driven evolution algorithms. Use when building self-improving agents, optimizing agent prompts and skills against benchmarks, or implementing automated agent evaluation loops.AI & Agents
SharePost on XLinkedIn
All trending reposRevenue-verified startups →