Maintainer of this skill?
by reacher-z
Open, source-backed catalog of computer-use agents: reviewed papers, GUI models, datasets, benchmarks, tools, and reusable JSONL/SQLite data.
# Add to your Claude Code skills
git clone https://github.com/reacher-z/Awesome-CUAGuides for using ai agents skills like Awesome-CUA.
See how Awesome-CUA compares with popular alternatives.
Awesome-CUA is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by reacher-z. Open, source-backed catalog of computer-use agents: reviewed papers, GUI models, datasets, benchmarks, tools, and reusable JSONL/SQLite data. It has 502 GitHub stars.
Awesome-CUA's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/reacher-z/Awesome-CUA" and add it to your Claude Code skills directory (see the Installation section above).
Awesome-CUA is primarily written in Python. It is open-source under reacher-z on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh Awesome-CUA against similar tools.
No comments yet. Be the first to share your thoughts!
Available after the catalog security scan passes.
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
A source-backed collection of papers, models, datasets, benchmarks and tools for agents that use computers.
English · 简体中文
| Project | Focus |
|---|---|
| WebsiteBench | Rebuild full-stack websites through browser exploration. |
| ClawBench | Evaluate browser agents on everyday online tasks. |
Search & filter · Research papers · Browse topics · Data & agents
Read the research survey — Methods, evaluation, human agency, and reusable research assets. English working manuscript with a versioned bibliography and reading register.
Find by task · Explore · Highlights · Recently reviewed · Use the data
Coverage ledger · Contribute · GitHub organization
Discovery is scheduled every two hours, at 00:07, 02:07, 04:07, 06:07, 08:07, 10:07, 12:07, 14:07, 16:07, 18:07, 20:07 and 22:07 UTC. Actual scan results · Schedule and safeguards.
For people, RAG systems, and agents — clone the complete catalog:
git clone https://github.com/reacher-z/Awesome-CUA.git
Agent instructions · RAG and agent guide · Complete JSONL · Retrieval documents
The goal is comprehensive coverage of public research and industry resources, with provenance that can be inspected. Papers, models, training data, evaluation environments, frameworks, infrastructure, products, and publishers belong in the same searchable catalog.
Global completeness is a goal, not a verified claim. We publish the source populations, review states, and remaining gaps. Bibliographic metadata verification is different from reviewing a resource's relevance or quality.
Browse by topic · Topic subsets and fingerprints
| Your task | Start here | What you can find |
|---|---|---|
| Find papers | Research library · Topics | Reviewed papers by year and topic, with original sources. |
| Choose a benchmark | Benchmark guide | Platforms, tasks, evaluators, and versions to compare. |
| Find training data | Training-data guide | Supervision, splits, count units, and data licenses. |
| Build an agent | Agents · Frameworks | Implementation entry points, supported platforms, and papers. |
| Use RAG or an agent | Agent guide · Data hub | Complete records or retrieval documents with review states and sources. |
| Trace papers to assets | Paper–asset connections | Code, datasets, and models linked through reviewed primary evidence. |
cd Awesome-CUA
python3 scripts/catalog.py search "grounding" --kind paper --status curated --limit 0 --json
python3 scripts/catalog.py search --kind benchmark --environment web --status curated --limit 0 --json
python3 scripts/catalog.py search --kind dataset --environment android --status curated --limit 0 --json
python3 scripts/catalog.py search --kind framework --status curated --limit 0 --json
python3 scripts/export_rag.py --output dist/rag-documents.jsonl
python3 scripts/catalog.py get arxiv:2404.07972 --related --json
1,674 curated resources · 1,202 reviewed papers and surveys · 103 discovery documents
Evidence through 2026-11-06T04:18:34Z. Only curated resources are counted above. Listed separately: 816 records awaiting review have bibliographic verification only, and 1,970 discovered candidates remain unresolved, not counting 166 links reviewed as non-resources or other addresses of indexed resources. 1,308 scope exclusions retain their original metadata in the archive. Counts overlap as explained below.
| Research and training | Engineering and industry |
|---|---|
| Research by topic · 1,183 reviewed papers · Full bibliography · 2,018 paper and survey records, including those awaiting review | Agents · 45 |
| Surveys · 19 · Learning resources · 2 | Frameworks · 79 |
| Datasets · 86 | Infrastructure · 14 |
| Benchmarks · 90 | Tools · 22 |
| Models and checkpoint releases · 98 | Products · 20 · Publishers and brands · 16 |
By platform: Web 649 · Desktop 334 · Windows 65 · macOS 30 · Linux 54 · Android 323 · iOS 22 · HarmonyOS 3 · Cross-platform 197
Coverage spans web, desktop, Android, iOS, Windows, macOS, Linux, and cross-platform work. Topics include grounding, screen understanding, planning, memory, long-horizon workflows, reinforcement learning, data generation, safety, robustness, and evaluation. Environment facets overlap; framework support is not assumed from a model's benchmark results.
The most-mentioned curated resources per category: the count is how many other cataloged resources mention each one in their reviewed first-party source (README, model or dataset card). It reflects connectedness inside this catalog, not quality, popularity, or benchmark performance. Every category page lists all reviewed entries. Surveys come first, newest first.
Start with the newest surveys · all 19 →
| Date | Survey | Environments |
|---|---|---|
| 2026‑10‑14 | Seeing, Maintaining, and Learning: The Evolution of Multimodal Agent Memory | unspecified |
| 2026‑09‑02 | Efficient GUI Agents: A Systems Survey of Observation, Memory, Action, and Runtime Optimization | cross-platform |
| 2026‑08‑30 | A Systematic Survey of Agentic Skills: Architecture, Lifecycle, and Security | cross-platform |
| 2026‑04‑17 | A Survey on Long-Term Memory Security in LLM Agents: Attacks, Defenses, and Governance Across the Memory Lifecycle | unspecified |
| 2026‑02‑12 | Agent Skills for Large Language Models: Architecture, Acquisition, Security, and the Path Forward | unspecified |
Benchmarks · all 90 →
| Resource | Mentions | Environments | Paper | License | Description |
|---|---|---|---|---|---|
| OSWorld | 21 | desktop, linux, windows, macos | 2404.07972 (2024) | code Apache-2.0 | Execution-based evaluation of agents performing tasks in real desktop environments. |
| WebArena | 15 | web | 2307.13854 (2023) | code Apache-2.0 | Self-hosted web environments and tasks for evaluating autonomous web agents. |
| BrowserGym | 9 | web | 2412.05467 (2024) | code Apache-2.0 | Unified interfaces for browser-agent tasks and benchmark environments. |
| ScreenSpot Pro | 8 | desktop, windows, macos, linux | — | code MIT | GUI grounding benchmark focused on professional high-resolution interfaces. |
| VisualWebArena | 7 | web | [2401.13649](https:/ |