Microsoft open-sourced assert, a framework that turns natural-language requirements into executable…
microsoft open-sourced assert, a framework that turns natural-language requirements into executable agent tests, because an estimated 99 percent of organizations deploy agents with no formal pre-production behavioral testing. the testing gap is the least discussed number in the agent stack.
nobody ships a service without staging, except the service that ships itself.
Context
Microsoft's blog post, dated 2 June 2026, introduces ASSERT (Adaptive Spec-driven Scoring for Evaluation and Regression Testing), an open-source MIT-licensed framework that converts natural-language behavior requirements into executable evaluations of AI models and agents. The blog says LLM judges were validated against human review on a sample across more than 10 behavior concepts, and cautions that it works best when behavior definitions are narrow and aggregate scores should be treated cautiously.
Open source and natural language to executable evaluations are first-party. The blog dates the release to 2 June 2026, so open-sourced is not new in September, and the note gives no date. The 99 percent of organizations with no formal pre-production behavioral testing was not found in the blog, project site or repository text, and the note's because is not stated by the sources, so it is unverified. No preview or GA label was found. The judge validation is Microsoft's own design claim. Nobody ships a service without staging is the author's opinion.
Related work
- ASSERT examples (GitHub) ↗Snippet only.
Watch next
- The source of the 99% estimate, and independent use of ASSERT.
Sources
- ASSERT: written intent to executable evals (Microsoft)commandline.microsoft.com
- ASSERT project siteresponsibleai.github.io
- responsibleai/ASSERT (GitHub)github.com
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 18 September 2026 at 17:47 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →