| <!doctype html> |
| <html lang="en"> |
| <head> |
| <meta charset="utf-8" /> |
| <meta name="viewport" content="width=device-width, initial-scale=1" /> |
| <title>AuraOne</title> |
| <link rel="stylesheet" href="style.css" /> |
| </head> |
| <body> |
| <main> |
| <h1>AuraOne</h1> |
| <p> |
| AuraOne is a human data company for AI teams and a workflow service for |
| enterprises. It supplies expert judgment, speech, physical-world data |
| and checked workflows. |
| </p> |
| <p> |
| <a href="https://auraone.ai">Website</a> | |
| <a href="https://auraone.ai/company/facts">Company facts</a> | |
| <a href="https://auraone.ai/open">Open tools</a> | |
| <a href="https://github.com/auraoneai">GitHub</a> | |
| <a href="https://www.linkedin.com/company/auraonecompany/">LinkedIn</a> | |
| <a href="https://x.com/AuraOneLabs">X</a> |
| </p> |
|
|
| <h2>Open tools</h2> |
| <p>All MIT licensed.</p> |
| <table> |
| <thead> |
| <tr><th>Tool</th><th>What it does</th><th>Source</th></tr> |
| </thead> |
| <tbody> |
| <tr> |
| <td>EvalKit</td> |
| <td>Python CLI for rubric validation and linting, deterministic scoring, reviewer agreement, judge calibration, drift and leakage checks, and evidence reports. Runs on files you own.</td> |
| <td><a href="https://github.com/auraoneai/open">auraoneai/open</a></td> |
| </tr> |
| <tr> |
| <td>rubric-spec</td> |
| <td>Validate, lint, diff, and convert portable AuraOne Rubric Schema v1 files.</td> |
| <td><a href="https://github.com/auraoneai/rubric-spec">auraoneai/rubric-spec</a></td> |
| </tr> |
| <tr> |
| <td>iaa-kit</td> |
| <td>Measure inter-annotator agreement with modern metrics, uncertainty, ordinal support, and missing-data handling.</td> |
| <td><a href="https://github.com/auraoneai/iaa-kit">auraoneai/iaa-kit</a></td> |
| </tr> |
| <tr> |
| <td>judge-bench</td> |
| <td>Probe judge behavior for position, verbosity, self-preference, paraphrase, anchoring, and calibration failures.</td> |
| <td><a href="https://github.com/auraoneai/judge-bench">auraoneai/judge-bench</a></td> |
| </tr> |
| <tr> |
| <td>lerobot-quality-gates</td> |
| <td>Check a LeRobot-style dataset's metadata, episodes, timestamps and video references before training or publication.</td> |
| <td><a href="https://github.com/auraoneai/lerobot-quality-gates">auraoneai/lerobot-quality-gates</a></td> |
| </tr> |
| <tr> |
| <td>Rubric Studio Open</td> |
| <td>Author, validate, preview, calibrate, diff, and export criterion-level AI evaluation rubrics.</td> |
| <td><a href="https://github.com/auraoneai/rubric-studio-open">auraoneai/rubric-studio-open</a></td> |
| </tr> |
| <tr> |
| <td>Agent Studio Open</td> |
| <td>Connect agent systems, inspect tool traces, replay failures, compare candidates, and export regression evidence.</td> |
| <td><a href="https://github.com/auraoneai/agent-studio-open">auraoneai/agent-studio-open</a></td> |
| </tr> |
| </tbody> |
| </table> |
|
|
| <h2>Packages</h2> |
| <p> |
| Python, on PyPI: |
| <a href="https://pypi.org/project/auraone-evalkit/">auraone-evalkit</a>, |
| <a href="https://pypi.org/project/rubric-spec/">rubric-spec</a>, |
| <a href="https://pypi.org/project/iaa-kit/">iaa-kit</a>, |
| <a href="https://pypi.org/project/judge-bench/">judge-bench</a>, |
| <a href="https://pypi.org/project/lerobot-quality-gates/">lerobot-quality-gates</a>. |
| </p> |
| <p> |
| JavaScript, on npm under the @auraone scope: |
| <a href="https://www.npmjs.com/package/@auraone/sdk">@auraone/sdk</a>, |
| <a href="https://www.npmjs.com/package/@auraone/agent-studio">@auraone/agent-studio</a>, |
| <a href="https://www.npmjs.com/package/@auraone/github-app">@auraone/github-app</a>, |
| <a href="https://www.npmjs.com/package/@auraone/aura-ide-kit">@auraone/aura-ide-kit</a>, |
| <a href="https://www.npmjs.com/package/@auraone/platform-contracts">@auraone/platform-contracts</a>, |
| <a href="https://www.npmjs.com/package/@auraone/proofline-oss">@auraone/proofline-oss</a>. |
| </p> |
|
|
| <h2>Datasets</h2> |
| <p> |
| Small synthetic tutorial datasets for the tools above. The rows are |
| hand-written. They are not expert-authored, not human-validated, and |
| not for training or benchmarking. |
| </p> |
| <ul> |
| <li><a href="https://huggingface.co/datasets/auraone/evalkit-tutorial-v0.1">evalkit-tutorial-v0.1</a></li> |
| <li><a href="https://huggingface.co/datasets/auraone/synthetic-multiturn-eval-v0.1">synthetic-multiturn-eval-v0.1</a></li> |
| </ul> |
|
|
| <p class="note"> |
| AuraOne (auraone.ai) is a different company from Aura, the |
| identity-protection service at aura.com. |
| </p> |
| </main> |
| </body> |
| </html> |
|
|
