Agentic Research Directory

Tool

AwesomeCUA: computer-use agent research catalog

AwesomeCUA documents a source-linked catalog of computer-use, GUI, browser and mobile agents, with research-paper discovery and JSONL/SQLite metadata downloads for people, RAG systems and agents. The repository is the static website and metadata snapshot, not the underlying models or datasets.

Workflow level n/a Source: Repository-documented Outcome: Not empirically evaluated here

Research applicability

A discovery and metadata input tool for researchers and AI-assisted literature workflows surveying computer-use agents or locating candidate benchmarks. It should guide primary-source reading rather than replace methodological assessment.

Getting started

Editorial, untested suggestion: read the catalog data and editorial-policy documentation, inspect one topic through the research-paper view, and trace candidate entries back to their primary sources. If importing JSONL or SQLite, preserve review-state and source fields and validate them before synthesis.

Maintenance

Dated documentation snapshot only; continued maintenance and availability not independently established.

License

README declares MIT for original code and annotations; linked assets retain separate licenses. License text and export contents not audited.

Workflow & human oversight

No automation level assigned. This catalog supports evidence discovery and agent ingestion; humans remain responsible for selecting sources and distinguishing bibliographic metadata from validated research.

Workflow level is not an evidence score; higher is not better evidence.

Approach

The README separates reviewed entries from metadata-only discoveries, offers links connecting papers to code and data, and documents machine-readable catalog exports. Bibliographic discovery is not experimental verification, and catalog counts are not counts of independent projects.

Limitations

README-only review: the website, download contents, individual papers, coverage claims and automation runs were not independently checked. No catalog completeness, annotation accuracy, retrieval quality or benchmark result is established. Upstream claims remain unverified by this review; metadata-only records must not be treated as reviewed evidence.

Reproducibility

Exact retrieved bodies, source URLs, UTC retrieval timestamps and SHA-256 envelopes saved. No upstream code or claimed outcomes reproduced. HEAD was mutable; no commit version pinned.

Adoption requirements

README describes a static site requiring no account or API key for browsing. Agent/RAG reuse still requires source-rights review, filtering by review status and checking upstream licenses. Downloaded metadata and annotations are not upstream training data, weights or papers.

Comments

No comments yet.

Add a comment

Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.

Submitting stores your comment text, the optional name and email, and a pseudonymised connection identifier used only for spam limiting (the raw IP address is not stored).