Legible Connect: AI Agent Knowledge Base | Integrate Your Website Content

AI agents are powerful. But they don't know your business.

You can ask Claude Code to write a blog post, draft a product spec, or generate customer-facing copy. The result is often impressive — and often wrong. Not because the AI is bad, but because it doesn't know your company. It doesn't know your messaging pillars, your product roadmap, your pricing structure, or the 47 blog posts you've already published on the topic.

Every team working with AI agents hits this wall. The agent is smart, but it's working from generic internet knowledge. So you end up copy-pasting context into prompts, uploading docs before every session, or just accepting that AI output needs heavy editing.

What if your AI agent could search your knowledge base?

That's what Legible Connect does. It exposes your Legible content — your website pages, blog posts, FAQs, documentation, uploaded files — as a live MCP (Model Context Protocol) server. AI agents that support MCP, like Claude Code and OpenAI Codex, can search, retrieve, and use your content as context automatically.

No copy-pasting. No file uploads. No stale context. Your agent pulls exactly what it needs, when it needs it.

How it works

Legible Connect works through the Model Context Protocol — an open standard for connecting AI agents to external data sources. Here's the flow:

1. You connect your site to Legible (if you haven't already). Legible indexes your content, chunks it for retrieval, and keeps everything current as you publish.
2. You add a single MCP config to your AI agent. It takes about 30 seconds — just paste your Legible endpoint URL and API key.
3. Your agent now has access to three tools: search_context to find relevant content, get_document to retrieve a full page, and list_sources to see what's available.

When you ask your agent to "write a blog post about our new pricing model," it calls search_context behind the scenes, finds your existing pricing page and related blog posts, and uses that real context to generate something accurate and on-brand.

Why this matters for your team

Content teams get AI that writes in your voice. When the agent has access to your messaging pillars, brand guidelines, and published content, it produces drafts that need tweaking, not rewriting.

Engineering teams get documentation-aware coding. Developers can ask questions about internal APIs, architecture decisions, and coding standards. The agent retrieves the relevant docs from Legible instead of guessing.

Support teams get accurate answers. The same knowledge base that powers your Content Chat and chatbot integrations is now available to any AI agent your team uses.

From content layer to context layer

When we launched Legible, the goal was to make your content legible to AI — discoverable, structured, and easy for AI systems to read. Legible Connect is the natural next step. Your content isn't just readable by AI crawlers anymore. It's usable by the AI tools your team works with every day.

The same content that powers your llms.txt, your Markdown delivery, and your Content Chat now powers your AI agents. One system, many outputs.

Get started

Legible Connect is available now for all Legible customers. Setup takes 30 seconds:

1. Go to your dashboard → Settings → API Keys and generate an API key.
2. Add the MCP config to your AI agent (see the setup guide for Claude Code, Codex, and others).
3. Start prompting. Your agent now retrieves your content automatically.

If you're not on Legible yet, start with a free account. Your content becomes AI-ready in minutes, and Connect comes included.