Make your product answerable to every AI agent
Your customers ask ChatGPT, Claude, and Perplexity before they ask you — and those agents shop your site on their behalf. HelpShelf gives them a live MCP server and an llms.txt surface, so they answer and sell your product correctly, and hand off to a human when it counts.
Can AI agents use HelpShelf?
Yes. HelpShelf gives every site a live MCP server and an llms.txt context bundle, so AI agents like ChatGPT, Claude, and Perplexity can search your product, get cited answers, and hand off to your team — powered by the same verified knowledge that runs your human-facing widget, inbox, and help center. It generates the surface for you, with a free tier to start.
Your customers ask AI before they ask you.
Search is becoming a conversation with an agent — and that agent decides what your product is.
- An AI agent answers for you — from a stale blog post, a competitor, or a hallucination.
- The agent gets a buyer to the brink, then has no way to capture the lead or reach you.
- You have a help center for humans, but nothing structured for the machines reading on their behalf.
- You cannot see what agents are asking, or where they get your product wrong.
From content to agent-ready in three steps
No SDK to integrate. No server to run.
Publish your agent surface
HelpShelf turns your content into an llms.txt bundle and a hosted MCP server automatically — no engineering required. Curated and published content only.
Agents query your product
ChatGPT, Claude, Perplexity, Cursor, or any MCP client can search your docs, ask questions, and pull product context — and get cited, accurate answers back.
Agents act, or hand off
When an agent hits something it cannot resolve, escalate_to_human routes it to your team — turning an autonomous AI session into a captured lead or ticket.
Everything an AI agent needs to get you right
Readable, cited, and actionable — with a human one tool-call away.
A live MCP server
Every site gets a hosted MCP endpoint with search_docs, ask, get_article, and get_product_context — so any AI agent can query your product directly.
llms.txt, generated for you
HelpShelf publishes llms.txt and a full llms-full.txt context bundle from your verified content, so agents read an accurate, curated version of your product.
Cited answers, not guesses
Agents get answers grounded in your sources with citations — drawn only from curated and published content, so they never invent features you do not have.
Human handoff built in
An escalate_to_human tool lets an external agent open a real conversation with your team — so an AI session can turn into a lead or a ticket, not a dead end.
One source of truth
The same knowledge that powers your widget, inbox, and help center answers the agents — update once, and every human and AI surface stays in sync.
Show up in AI answers
Being structured, cited, and agent-readable is how you get surfaced by ChatGPT, Claude, and Perplexity — answer engine optimization, handled by default.
Speaks to every agent
One MCP + llms.txt surface, read by the assistants your customers already use
Frequently asked questions
Yes. HelpShelf gives every site a live MCP server and an llms.txt context bundle, so AI agents like ChatGPT, Claude, and Perplexity can search your product, get cited answers, and hand off to your team — powered by the same knowledge as your human-facing widget. There is a free tier to start.
MCP (Model Context Protocol) is the standard that lets AI agents call tools and pull context from external systems. A HelpShelf MCP server exposes read tools (search_docs, ask, get_article, get_product_context) plus escalate_to_human, so an agent acting for your customer can answer accurately about your product and reach a human when needed.
Any tool that supports MCP or can read a published llms.txt — including ChatGPT, Claude, Perplexity, and AI code editors like Cursor and Windsurf. HelpShelf also adds Ask-with-my-AI buttons that deep-link your help content into those assistants.
No. Point HelpShelf at your content and it generates the llms.txt bundle and hosts the MCP endpoint for you. There is no SDK to integrate and no server to run — the agent surface goes live alongside your widget.
A chatbot answers humans on your site. The agent surface makes your product readable and actionable by the AI agents your customers already use elsewhere — so when someone asks ChatGPT about your product, the answer is accurate, cited, and can route a real lead back to you.
Yes. Answer engine optimization and generative engine optimization reward content that is structured, cited, fresh, and machine-readable. Publishing a curated llms.txt and an MCP surface is exactly the substrate AI answer engines look for, which improves how often — and how accurately — you are cited.
Support for the humans, too
The same agent answers your customers in the widget, inbox, and help center — and adds support to the app you built.