Agent Score: can AI agents read your docs?
Can AI agents read your docs?
Run it locally or in CI.
Agent Score is a hosted scan. For private docs or a check in your pipeline, use afdocs, the open-source library it's built on. MIT-licensed, by Dachary Carey.
$ npx afdocs@latest check https://docs.example.com \
--skip-checks llms-txt-coverage --format scorecard
Overall Score: 82 / 100 (B)
Content Discoverability 74 / 100 (C)
Markdown Availability 63 / 100 (D)
Page Size and Truncation Risk 92 / 100 (A)
…Leaderboard
Your docs have a second reader.
Developers send Claude Code, Cursor, and ChatGPT to read your docs before they do. An agent that cannot parse a page does not complain. It answers with less.
Want an agent to audit the content, not just the plumbing?
Agent Score only runs deterministic checks: is there an llms.txt, is there a markdown version of each page, does a page fit in one fetch. It cannot tell you whether your docs answer the questions agents get asked. For that, sign up for Promptless and run a content strategy audit on your docs. Free 14-day trial, no credit card.