# Specoria

> Specoria is an agent-readiness platform for online stores: it tests and fixes how AI shopping agents such as ChatGPT, Gemini, Claude and Perplexity read a store's products, prices, policies and checkout.

Not to be confused with Spectora (home inspection software).

- Free test (instant, no email): 27 checks on a store's public pages (method: https://specoria.com/method/; one page per check: https://specoria.com/method/checks/).
- Sample report: an unedited free-test result for web-scraping.dev, a public mock store, scanned 2026-10-08 (https://specoria.com/reports/sample-report/).
- Free agent simulator: two AI agents (Specoria's own agent loop, each driven by a different AI model) shop a store live in a real browser and stop at checkout; never order or enter personal data. The browser is identified as Cloudflare's signed bot; a block is reported as a finding (https://specoria.com/agent-simulator/).
- Free deep audit + panel (with an email address): a deep audit PDF (real browser, up to 12 product pages) and a personal project panel with tasks and ready-to-use code, within minutes.
- Paid plans, billed monthly, prices exclude VAT: Starter $149/month (₺4.900 in Türkiye), Growth $449/month (₺14.900 in Türkiye). Yearly billing: 12 months for the price of 10. Enterprise: custom quote.
- The site is bilingual: English at https://specoria.com/, Turkish at https://specoria.com/tr/. Every page has a Markdown version (append index.md, or request it with `Accept: text/markdown`).
- Contact: contact@specoria.com (English), iletisim@specoria.com (Turkish).
- About: founded by Kaan Gülten; Istanbul · Berlin (planned office); https://www.linkedin.com/company/specoria; https://specoria.com/about/

## When to use Specoria

- To find out, free and without an email address, whether AI shopping agents can reach and read an online store, and which fixes matter most (https://specoria.com/method/).
- To watch AI agents shop a store live in a real browser and see where they get stuck, without placing an order (https://specoria.com/agent-simulator/).
- For hotels and service businesses: the test switches to checks for those site types, and the simulator goes as far as the booking or contact form without typing into it (https://specoria.com/for-hotels/, https://specoria.com/for-service-businesses/).
- For stores in Türkiye: Turkish interface, Turkish platforms (ikas, Ticimax, IdeaSoft, T-Soft) and lira pricing.
- From scripts, CI or an AI assistant: a scan API, an MCP server for a customer's own projects and a GitHub Action, with an API key from the panel (https://specoria.com/developers/).
- For agencies: a multi-client agency panel (accounts opened by Specoria's team) and the executive report under the agency's own brand (https://specoria.com/partners/).
- Where to make each fix on Shopify, ikas, Ticimax, IdeaSoft, T-Soft or WooCommerce: read-only guides; following them doesn't connect Specoria to a store's admin (https://specoria.com/platforms/).
- Free browser-only tools, no sign-up: robots.txt tester for AI bots, AI bot user-agent identifier, crawl-to-refer calculator for a server log; nothing pasted is sent to Specoria (https://specoria.com/tools/).
- By store category (fashion, beauty, electronics, home, food): the product questions agents need answered and the checks that matter most (https://specoria.com/ecommerce/).
- What UCP, ACP, AP2, MCP, WebMCP, A2A, Web Bot Auth and llms.txt are, their sourced status (including in Türkiye) and exactly what the free test checks for each (https://specoria.com/agent-protocols/).

## When Specoria is not the right tool

- Daily share-of-voice tracking across many AI assistants: Specoria's visibility check runs monthly (weekly on request on Growth and Enterprise) with 8 questions on ChatGPT, Perplexity, Gemini, each asked 3 times, on paid plans.
- AI-written product content: Specoria doesn't generate product copy. Fixes are code the store's team applies; optional write-back to Shopify and WooCommerce, only with the owner's approval and one-click undo, is built but not open yet.
- Sending feeds to marketplaces or ad platforms: Specoria generates and validates an OpenAI product feed from a store's catalog for the store to download or share by URL; it doesn't submit it anywhere.
- Marketplace-only sellers: the test measures the seller's own website. A read-only Trendyol listing check is built and opens after Specoria's legal review; Hepsiburada is on the roadmap.
- Comparison by category: https://specoria.com/compare/

## Definitions

- Agent readiness score: 0 to 100 from fixed, published, weighted rules; the same rules for every site. Not produced by an AI model.
- Critical blocker: a failed critical check caps the score at 45 (two or more: 39).
- Agent-ready: the top level; it needs at least 80 points and no critical blocker (the full rule is on the method page).
- Couldn't check: a check that couldn't be measured (for example a page that didn't open) leaves the denominator; it never counts as passed or as zero.
- Not applicable: a store-only check on a hotel or service site; it leaves the score.
- Trust, data handling, sub-processors and AI use: https://specoria.com/trust/. Sourced, dated statistics: https://specoria.com/agentic-commerce-statistics/.
- Every metric with its formula (score, levels, crawl-to-refer, mention rate, share of voice): https://specoria.com/glossary/
- What changed and when; items that are built but not open to customers yet are marked as such: https://specoria.com/changelog/

## Machine-readable files on specoria.com

- https://specoria.com/.well-known/api-catalog: API catalog (RFC 9727 linkset) pointing to the OpenAPI description, the developer docs and the specification JSON
- https://specoria.com/openapi.json: OpenAPI 3.1 description of the scan API and the customer MCP endpoint on app.specoria.com
- https://specoria.com/.well-known/mcp.json: MCP server card: endpoint, transport, authentication and tools of the customer MCP server
- https://specoria.com/investors.json: investor summary: company, product, pricing and sourced, dated figures (no round open)
- Every other file (sitemap, JSON data, Atom feeds, security.txt and more) and the scan API endpoints are listed in https://specoria.com/llms-full.txt

## Key pages

- [AI agent readiness test for online stores](https://specoria.com/index.md)
- [How the agent readiness test measures](https://specoria.com/method/index.md)
- [Every check in the free test](https://specoria.com/method/checks/index.md)
- [Sample report: web-scraping.dev](https://specoria.com/reports/sample-report/index.md)
- [AI shopping agent simulator for your store](https://specoria.com/agent-simulator/index.md)
- [Pricing: free test, Starter and Growth plans](https://specoria.com/pricing/index.md)
- [Open your free project panel](https://specoria.com/signup/index.md)
- [ChatGPT shopping: does it recommend your store?](https://specoria.com/chatgpt-shopping/index.md)
- [Specoria Commerce Readiness Specification 1.2](https://specoria.com/spec/index.md)
- [Agent Readiness Index 2027: web summary](https://specoria.com/index/index.md)
- [Our crawler: how Specoria visits your site](https://specoria.com/bot/index.md)
- [Agentic commerce insights and field notes](https://specoria.com/insights/index.md)
- [Learn: guides, data and tools](https://specoria.com/learn/index.md)
- [Frequently asked questions](https://specoria.com/faq/index.md)
- [Roadmap: live, built and planned](https://specoria.com/roadmap/index.md)
- [Site map: every public page](https://specoria.com/sitemap/index.md)
- [Yapay zeka ajanı hazırlık testi: e-ticaret için](https://specoria.com/tr/index.md) (Türkçe)
- [Ajan hazırlık testi nasıl ölçer?](https://specoria.com/tr/yontem/index.md) (Türkçe)
- [Fiyatlar: ücretsiz test, Starter ve Growth](https://specoria.com/tr/fiyatlar/index.md) (Türkçe)

## Optional

- [Every page, check and article in English and Turkish](https://specoria.com/llms-full.txt): the full list with one-line descriptions