Blume Hits Product Hunt's #1 Spot With Docs Framework

Blume took Product Hunt's number one spot on August 16 with a documentation framework for AI agents and human readers, per Product Hunt.

That's the pitch for teams maintaining docs: the AI surface ships built in, not bolted on later. An MCP server, llms.txt, and per-page Markdown come standard, which is what separates Blume from being just another static site generator, per Product Hunt.

Point the CLI at a folder of Markdown, and it builds a production site with zero setup. Behind the scenes, it runs a hidden Astro and Vite project. The upshot, though Product Hunt doesn't spell it out, is that an AI agent can read the docs directly instead of scraping rendered HTML.

Run blume eject, and it hands over a standalone Astro app instead of locking teams into the CLI. Quiver has already migrated its docs off Mintlify to Blume.

The framework also ships more than 30 MDX components, local search through Orama, and an OpenAPI reference rendered with Scalar, all under an MIT license. The code first shipped in mid-July, and the Product Hunt launch and number one finish came weeks later, on August 16.