We ran the loop by hand.Then we built Lyra.

Connect her to your repo and she reads where your posts live, your frontmatter, your slug rules, and how you sound. She researches live search data for winnable topics and dedupes them against everything you have published, so she never cannibalizes your own rankings. Every external link gets fetched and confirmed or dropped, and every claim gets checked against current sources. Each draft is scored out of ten across content, SEO, technical accuracy, readability, and linking, and she rewrites it until it clears the bar.

The result lands as a labeled pull request with a banner, and she tags you to merge. She runs on your own Anthropic key, encrypted at rest and billed at cost, never marked up.

Before Lyra, we grew a developer product whose search presence was a flat line near zero. Paid acquisition worked until we stopped paying, and social spikes faded in a week. Blog content was the one channel that compounded: a post that answered a real question kept pulling readers months after we wrote it. So we wrote one post a week, then two. Around month four a few posts cracked the first page for long-tail queries, and by the time we looked up the blog was doing millions of monthly search impressions.

The cost was the grind. Every week meant researching topics, deduping against old posts, verifying every statistic and link, wiring internal links, generating a banner, and opening the pull request. It worked, and it was ninety percent rules. That is the moment a founder either hires a content team or builds a machine. We build machines, so we built Lyra to run the loop for any team with a blog in a repo. The full story is in why we built Lyra.