# Serge

> Serge tells e-commerce sites whether users behind AI agents can actually find their products and buy them. When a customer asks Claude, ChatGPT, or Operator to buy something on a site, the agent either succeeds or gives up. Serge measures which — and what to fix when it's the second one.

## What Serge does

Serge scans e-commerce sites and scores how easy it is for an AI agent (acting on behalf of a human shopper) to find a product and place it in a cart. A single number from 0 to 100. Deterministic — no LLM calls at scan time. Each finding includes a specific fix.

Serge occupies the middle layer of a three-layer problem:

- **Upstream (visibility):** Does ChatGPT mention you in its answers? That's GEO. Covered by Athena, Profound, Scrunch, Peec, Otterly, and others.
- **Middle (arrival + traversal):** Can the agent actually find and buy your products once it lands on your site? **This is Serge's layer. The category was empty until Serge.**
- **Downstream (attribution):** What share of revenue came from LLM-mediated sessions? That's attribution tooling — Dreamdata, HockeyStack, Bizible.

Serge is built for e-commerce Product Directors and Front End / Engineering Leads at mid-to-large e-commerce sites, with Swiss and DACH enterprise retailers as the first design-partner cohort.

## Products

- **Serge Scanner** (https://serge.ai/scanner): Free deterministic scan of any e-commerce domain. Paste a URL, get a score plus a list of findings and fixes. No signup. The structural-readiness wedge.
- **Agent Journey Tests** (https://serge.ai): The paid product. Serge runs real AI agents — Claude, ChatGPT, Operator, Perplexity, Gemini — against your live storefront with a buying task ("find a 13-inch laptop under $1200 and add it to cart"). Captures the full session — reasoning, DOM events, screenshots, the exact failure point — and renders a shareable journey-test report with a paste-ready fix snippet per failure. Free deterministic scan, sponsored replays for early evaluators, then Pro at CHF 159/mo.
- **Live demo** (https://serge.ai/preview/agent-journey-tests/live-demo): Public preview of a journey-test report. The meta variant: an agent shopping serge.ai itself, asked to pick the right Serge plan. Five steps, all green.

## Where Serge fits vs other tools

Comparison pages with side-by-side coverage tables + FAQ:

- **vs Profound** (https://serge.ai/compare/profound): Visibility (does ChatGPT mention you?) vs conversion (can it buy from you?). Different layers, same buyer.
- **vs Hotjar** (https://serge.ai/compare/hotjar): Human session replay vs agent journey tests. Different traffic types; both useful.
- **vs Contentsquare** (https://serge.ai/compare/contentsquare): Enterprise human DX analytics vs agent journey tests.
- **vs FullStory** (https://serge.ai/compare/fullstory): Broad human session capture vs agent journey tests.
- **vs Quantum Metric** (https://serge.ai/compare/quantum-metric): Felix AI detection (flags agent-like traffic) vs agent task completion (tests whether agents can buy).

## What the scanner measures

One question: **can a user behind an AI agent find a product on this site and place it in a cart?**

The scoring methodology is being rebuilt from the ground up (as of April 2026). The research direction covers the structural conditions an agent needs to complete the journey:

- Can the agent reach the site? (bot protection, robots.txt, WAF posture)
- Can the agent find products? (sitemap, navigation semantics, product URL discoverability)
- Can the agent parse product data? (schema.org Product, Offer, Availability, clear pricing, variant metadata)
- Can the agent interact with the page? (real buttons vs. `<div onClick>`, accessible names, ARIA roles)
- Can the agent add to cart? (cart state exposure, variant selection mechanics, inventory visibility)

The final check list and scoring rubric are the output of a research + validation workstream. Published openly at https://serge.ai/methodology when locked.

## API

- Endpoint: `POST https://serge.ai/api/scan` — initiate a scan
- Endpoint: `GET https://serge.ai/api/scan/{id}` — retrieve scan results
- Authentication: Bearer token via `Authorization: Bearer sk_serge_...` header
- Anonymous: 10 scans/hr per IP. Authenticated: 60 scans/hr per workspace.
- Create secret keys at https://serge.ai/member/api-keys
- OpenAPI spec: https://serge.ai/openapi.json

## Extended documentation

For comprehensive documentation including scoring details, methodology, and API examples, see [llms-full.txt](https://serge.ai/llms-full.txt).

## SergeBot (crawler)

SergeBot is the crawler that powers the scanner. It identifies itself as `SergeBot/1.0` and only requests publicly accessible files — product pages, structured data, robots.txt, sitemap. It does not scrape, index, or store page content.

- Bot documentation: https://serge.ai/bot
- Contact: bot@serge.ai

## Company

Serge is built by Superstellar LLC, based in Zug, Switzerland.

- Website: https://serge.ai
- Contact: hello@serge.ai