About Dedica

Dedica checks whether an AI agent, such as ChatGPT or Claude, can actually find a website, read its pages and act on what it finds there. Most sites were built for human visitors clicking through a browser; an increasing share of traffic is now an agent working on someone else's behalf, and the two audiences need different things from the same pages.

What the check looks at

A scan reads the site the way an agent would: whether it is blocked by bot protection, whether its pages carry readable content without running JavaScript, whether it publishes the files agents look for first (an llms.txt index, markdown fallbacks of key pages, structured data, an API an agent could call directly), and whether it offers a way for an agent to complete a task rather than only read about it. The result is a score split across four layers: Discovery (can an agent find and identify the site at all), Access (can it get past the front door and read the content), Usability (can it act once it is in), and, where relevant, Payments.

Why a score instead of a checklist

Individual checks move; a static checklist goes stale. Dedica re-runs the same checks on demand, so a score reflects the site as it is right now, not as it was when someone last audited it by hand.

What you get

The free report shows the score and a sample of the findings behind it. The paid fix kit, a one-time purchase per domain, unlocks every finding together with a ready-to-run instruction file for each one, written to be handed to a coding assistant such as Claude Code, Codex, Cursor or GitHub Copilot rather than read and re-typed by a person. See pricing for the current price, and developers for the API, MCP server and other endpoints an agent can call directly.

Who runs this

Dedica is a small, independently run product. See Terms of Service and Privacy Policy for the legal and data-handling details.