agenticweb
Health Warn
- License — License: Apache-2.0
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 5 GitHub stars
Code Warn
- network request — Outbound network request in bench/bench.ts
Permissions Pass
- Permissions — No dangerous permissions requested
No AI report is available for this listing yet.
Make any website you own easy for AI agents to read and use: llms.txt, Markdown, MCP, WebMCP and safe actions.
AgenticWeb
Make your website easy for AI agents to read and use.
AI agents now shop, compare and book for people. But websites are built for eyes, so agents get lost in menus, scripts and pop-ups, read far more than they need, and can't safely press "Book" or "Buy".
AgenticWeb reads a website you own and builds an agent version of it next to the real one: the same content, in clean formats agents understand, with safe tools for actions like booking a table or starting a checkout. Your site for people stays exactly as it is.
https://github.com/user-attachments/assets/800ed286-fa7c-4ea2-8629-28ea9f93d51d
How much it helps
We gave agents 18 tasks on three demo sites (a restaurant, a shop and a docs site), such as "What time do you close on Sunday?", "Is there a vegan pizza?" and "Put a blue hoodie, size M, in the cart".
| Tasks done | Requests | Text read | |
|---|---|---|---|
| Normal website | 17 of 18 | 47 | about 29,300 tokens |
| With AgenticWeb | 18 of 18 | 19 | about 4,800 tokens |
Agents read 84% less and made 60% fewer requests, and only the agent version could put items in a cart. That's against a best-case reader that stops the moment it sees the answer; real agents usually do worse.
The demo sites were built alongside AgenticWeb, so treat this as a demonstration until it's repeated on real sites. Full results are in bench/results-heuristic.md, and you can rerun them with npm run bench.
Get started
You need Node 22.18 or newer. The npm package is called agenticweb-cli; if you install it (npm install -g agenticweb-cli), the command is just agenticweb.
npx agenticweb-cli init https://mysite.com
npx agenticweb-cli build
npx agenticweb-cli review
npx agenticweb-cli serve --edge
- init starts a project and asks you to prove you own the site (a DNS record or a small file).
- build reads your site and pulls out what matters: pages, products, menus, hours, FAQs and policies.
- review opens a page where you choose what agents can see and which actions they can use. Everything starts off.
- serve --edge shows your site with the agent version added, at
http://127.0.0.1:8787.
Do I need an AI key?
No. AgenticWeb works fully without one. The benchmark above was run with no AI key at all.
What an AI provider is for: it's the AI that AgenticWeb can call on for a few optional extras. The core work never depends on it: building the agent version, the safety checks, bookings and checkout all run on AgenticWeb's own rules.
| Without a key | With a key | |
|---|---|---|
| Reading your site | Built-in rules, good for well-built sites | Smarter on messy pages (and PDFs, with Claude) |
| Answering agents' questions | The matching facts, with sources | A short written answer, with sources |
| Testing your site as an agent | Quick checks with set answers | A real AI agent tries each task |
| Questions your site can't answer yet | The question, for you to answer | A draft answer from your site's own facts |
Pick any provider: Claude (the default), OpenAI, DeepSeek, Gemini, or a free model running on your own computer (Ollama or LM Studio). It's one line in the config, and the guide shows how. Your key stays on your computer or server, and requests go straight to the provider you chose, never to us.
Put it on your site
Recommended: edge mode. AgenticWeb runs in front of your site on your own domain. Nothing gets installed on your site, so it works with WordPress, Shopify themes, Webflow, custom code or anything else.
npx agenticweb-cli deploy --target edge
This prepares a Cloudflare Worker for your domain. Not on Cloudflare? Use --target docker or --target node and run it in front of your site. You can also host the agent version on its own address, like agents.mysite.com. The guide covers every option.
What agents get
| What it is | |
|---|---|
/llms.txt |
A short summary of your site for AI models |
| Markdown pages | Any page, as clean text, when an agent asks for it |
| Product and business data | Prices, stock, hours and FAQs in the standard schema.org format |
/.agent/ask |
Ask a question in plain words and get an answer with sources |
| MCP server | Your site as tools for ChatGPT, Claude, Gemini, Cursor and other AI apps |
| WebMCP | The same tools for AI agents built into browsers |
| UCP and ACP checkout | The checkout protocols shopping agents use, for any store, not only Shopify |
Check that it works
- Test it like an agent.
npx agenticweb-cli testtries your real tasks, like "book a table for 4 on Friday", on both your site and the agent version (with an AI key, a real AI agent does them). You get an Agent Readiness score, a badge, and a warning when a task that used to work breaks. - Prove it's the same story. Every build checks each fact (prices, hours, contact details) against your live pages and signs the result with your key. Anyone can check it with
npx agenticweb-cli check <agent-branch-url>. - Fill the gaps.
npx agenticweb-cli suggestlists questions agents asked that your site couldn't answer, with a draft answer for each if you have an AI key. You approve them and add them to your site, and agents pick them up from there.
Works with
Shopify, WooCommerce, Stripe, Cal.com, any API with an OpenAPI description, any HTTP endpoint, and plain HTML forms (opt-in).
Safe by default
- Only the owner can build it. You have to prove you own the site first.
- Same facts for people and agents. The agent version is a different format of the same content, never a different story. A signed parity report lists any fact that doesn't match your pages. Agents get Markdown only when they ask for it.
- You decide who can do what. Agents that sign their requests (Web Bot Auth) or use a key you gave them are verified. You can let verified agents book while others only read.
- Actions are off until you turn them on. Nothing final happens without the check you chose: the customer confirms by email, you approve, or the customer pays on your own checkout.
- No card numbers, ever. Payments go through your checkout or your payment provider.
- Your data stays yours. AgenticWeb sends nothing to us. What your servers log, such as IP addresses and booking details, is listed in the privacy section of the guide so you can cover it in your privacy policy.
- Website text is treated as data, not instructions. AgenticWeb never follows text it reads on a page, and it tells agents not to either.
More in SECURITY.md.
Use it from your AI assistant
- Any coding assistant (Claude Code, Codex, Cursor, Gemini CLI and others) can run the
npx agenticweb-clicommands for you. - Any MCP app (ChatGPT, Claude Desktop, Cursor, VS Code and others): add an MCP server with the command
npxand the arguments["agenticweb-cli@latest", "mcp"]. - Claude Code plugin: run
/plugin marketplace add adibfarah/agenticweb, then/agentify https://mysite.comfor a guided walkthrough.
Learn more
- Guide: every command and setting
- Format spec: the files and endpoints, so any tool can read or produce them
- WordPress and Shopify setup
- Contributing and the original plan
License
Apache 2.0. The npm package bundles open-source libraries under their own licenses (MIT, BSD, ISC, Apache 2.0 and Unlicense), listed with their full texts in THIRD_PARTY_NOTICES.md inside the package.
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found