Awesome Graph Engineering v1.4.2

This release attributes the archived dataset to its author's ORCID iD, 0000-0002-5728-9672, on Zenodo and in CITATION.cff. The 620 resources are unchanged from v1.4.0.

Citation

He, C. (2026). Awesome Graph Engineering: A Field Guide, Dataset, and Interactive Atlas for Programmable AI-Agent Organizations (Version 1.4.2) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.23617107

The DOI above always resolves to the latest version. Please keep the title as written so that citations to every version are grouped together.

Awesome Graph Engineering v1.4.1

This release archives the catalog on Zenodo so that it has a citable DOI. The 620 resources are unchanged from v1.4.0; see the v1.4.0 notes for what changed since July.

Citation

He, C. (2026). Awesome Graph Engineering: A Field Guide, Dataset, and Interactive Atlas for Programmable AI-Agent Organizations (Version 1.4.1) [Data set]. GitHub. https://github.com/ChaoYue0307/awesome-graph-engineering

Please keep the title as written so that citations to every version are grouped together. Once Zenodo publishes the record, its DOI is added to every citation block.

Changes

• .zenodo.json describes the archive as a CC0 dataset with the same title and creator as CITATION.cff, and is generated from it.
• Citation blocks, the APA line, and the site's dataset metadata pick up the DOI automatically once it is recorded in CITATION.cff.

Awesome Graph Engineering v1.4.0

The catalog grows from 118 to 620 curated resources across 11 sections and 9 design layers, including 298 papers and preprints, with the scope, evidence labels, and boundary tests unchanged.

What's new

• Scope and definition. The working definition now cites the first survey to adopt the label, Graph Engineering in the Era of LLM Agents (arXiv 2608.21156), and maps its task, agent-coordination, and runtime-state views onto the org graph and run/work graph.
• Research through October 2026. New work on when decomposition pays, topology selection under cost, authority boundaries between supervisors and subagents, agent-to-agent failures across principals, deterministic handoff validation, budget conservation across delegation, and privacy leakage along topology.
• Benchmarks. Coordination-specific benchmarks that score edges, dependencies, memory isolation, and collaboration gain separately from task success, including alem, CoCoBench, ForestBench, MASTraceBench, CoMemBench, and AsynCodeBench.
• Critiques and limits. Pre-registered and controlled results where multi-agent systems lose to a single agent, plus a noise-floor protocol for judging small coordination gains.

Citation

The title is now identical in CITATION.cff, GitHub's "Cite this repository" button, every BibTeX block, and the site's dataset metadata:

He, C. (2026). Awesome Graph Engineering: A Field Guide, Dataset, and Interactive Atlas for Programmable AI-Agent Organizations (Version 1.4.0) [Data set]. GitHub. https://github.com/ChaoYue0307/awesome-graph-engineering

Please keep the title as written so that citations to every version are grouped together.

Maintenance

• Weekly audits for archived or renamed repositories, undisclosed arXiv withdrawals, stale link-check exclusions, and drift in the Hugging Face mirror.
• Generated counts, translated share copy, and citation metadata are derived from source files and checked in CI.

Assets

The release includes the dataset as JSONL and CSV, the JSON Schema, a manifest, a packaged archive, and SHA-256 checksums.

Awesome Graph Engineering v1.3.0

Full Changelog: v1.1.0...v1.3.0

Awesome Graph Engineering v1.1.0 — Expanded Evidence Atlas

What’s new

• Expanded the evidence atlas from 51 to 80 curated resources across research, standards, tools, benchmarks, case studies, and critiques.
• Strengthened the evidence-grounded definition, scope boundary, taxonomy, and inclusion methodology for graph engineering.
• Added seven localized entry points: Simplified Chinese, Spanish, French, German, Japanese, Korean, and Brazilian Portuguese.
• Refined the interactive atlas and graph visuals, with stronger mobile, desktop, landscape, overflow, and reduced-motion behavior.
• Kept the canonical 13-field CC0 dataset synchronized across JSONL, CSV, the website, the README, and the Hugging Face mirror.

Downloads

• resources.jsonl — canonical, streaming- and diff-friendly dataset snapshot.
• resources.csv — spreadsheet- and analysis-friendly dataset snapshot.
• awesome-graph-engineering-v1.1.0-data.zip — datasets, documentation, definition, methodology, taxonomy, citation metadata, and license.
• SHA256SUMS — SHA-256 integrity checks for every uploaded data asset.

GitHub also provides source-code archives for this tag.

Citation and license

Suggested citation: He Chaoyue (2026). Awesome Graph Engineering (Version 1.1.0) [Dataset]. GitHub. Repository

Original metadata, schema, summaries, documentation, code, and visuals in this repository are dedicated to the public domain under CC0 1.0. Linked third-party works retain their respective rights.

Full comparison: v1.0.0...v1.1.0

Awesome Graph Engineering v1.0.0

First public release of the evidence-grounded Graph Engineering field guide, interactive atlas, multilingual front doors, and synchronized 51-resource dataset. Includes JSONL and CSV snapshots for reproducible citation.