mirror of
https://github.com/supabase/supabase.git
synced 2026-10-05 09:25:06 +03:00
d47747477d
## I have read the [CONTRIBUTING.md](https://github.com/supabase/supabase/blob/master/CONTRIBUTING.md) file. YES ## What kind of change does this PR introduce? - Adds a new blog post announcing the open-source `supabase/evals` benchmark, published at `/blog/introducing-supabase-evals` (date 2026-07-31) - Adds the post's images (`og.png`, `thumb.png`, and an inline benchmark-results chart) under `apps/www/public/images/blog/introducing-supabase-evals/` ## What is the current behavior? N/A. No existing post for this launch. ## What is the new behavior? - New MDX post `apps/www/_blog/2026-07-31-introducing-supabase-evals.mdx`, author `matt_rossman`, category `product` - Post covers what Supabase Evals is, why we built it, how the benchmark and regression suites work, key findings, and where agents struggle ## Additional context - Opened as a draft. Content is still under review in Notion, so this is not ready to merge yet. - Part of the Introducing Supabase Evals launch (Tier 2, target 2026-07-31). - Images are placeholders pending final assets from Brand Design if needed. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Documentation** * Added a new blog post introducing Supabase Evals, an open-source benchmark for evaluating AI coding agents on real Supabase tasks. * Described how evaluations run, how scoring works, retry behavior, update cadence, and common failure areas. * Included links to explore results and shared plans for expanding scenarios, improving scoring rigor, and adding feedback capabilities. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Ana <ana1337x@users.noreply.github.com> Co-authored-by: Matt Rossman <22670878+mattrossman@users.noreply.github.com>