User Manual

Everything you need to run Agentimus as a site owner — no code required. These pages walk you from installing the plugin to getting every feature working, in plain language.

Agentimus does two things: it makes your site legible and citable to AI, and it lets the AI tools you already use operate your site over an opt-in MCP server. Both sides are covered below.

New here? Read them roughly in order:

Overview — what Agentimus does and why it helps.
Installation — get set up in a couple of minutes.
Dashboard & activity — your AEO/GEO score, AI traffic, and what agents are doing.
Agent Access — who authenticates to and acts on your machine surface (approved assistants, connection tokens, application passwords, abilities), and what was refused.
Readiness — the checks behind your score, plus the per-page Optimize worklist, with one-click fixes.
Machine-readable outputs — llms.txt, Markdown, discovery docs, sitemap.
Structured data — the Schema.org (JSON-LD) your pages emit, plus the one-line AI description behind each page’s summary.
Identity — tell AI who you are and what you do.
Topics for AI — say what each page is about.
Write with AI — let your configured AI draft the descriptions, topics and fixes.
Agent preview — see exactly what an agent receives.
MCP server — connect Claude, Claude Code, Cursor, ChatGPT or Codex to your site by approving them: let them read your reports, and (opt-in) draft, edit and publish content and apply fixes.
AI access & crawlers — who may crawl, and how.
Exposure & hardening — optional privacy tightening.
AI Visibility — track whether AI engines cite you.
Caching & CDNs — keeping a page cache out of the way of the AI files.
Privacy & data — what leaves your server (almost nothing).
FAQ — quick answers to common questions.

Use the sidebar to jump to any page.

User Manual

Table of contents