doclight — ai Readiness Report
partisepeti.com
Pages analyzed (10)
llms.txt found
AI READINESS SCORE
0 /100
Partisepeti is a Turkish e-commerce marketplace selling party supplies and decorations (balloons, door decorations, banners, costumes, piñatas, themed party sets) for birthdays, baby showers, Halloween, Valentine's Day, and other special occasions, with 5 physical stores in Istanbul. The crawled pages are consumer-facing product and category pages plus llms.txt files, with prices in Turkish Lira; there is no programmatic API despite the pages being labeled 'api'.
Dimension Breakdown
Discoverability
70
The site is clearly identifiable as a Turkish party-supplies marketplace and offers llms.txt, llms-full.txt, and a product sitemap for crawling.
Comprehension
65
What the business sells is clear from product/category pages and the llms.txt summary, though entirely in Turkish and oriented to human shoppers.
Setup clarity
10
There is no account/API setup flow, authentication, or developer onboarding since this is a retail storefront, not a programmable product.
Documentation
25
llms.txt provides useful structural guidance and URL patterns, but there is no technical/API documentation for an agent to act programmatically.
Pricing clarity
60
Individual product prices and shipping thresholds are clearly stated in TL, but there is no machine-readable catalog or structured pricing endpoint.
Integration examples
5
Despite the 'api' labels, no OpenAPI spec, code samples, or integration methods exist; pages are only HTML product listings.
Agent Journey
✓
Discover
✓
Understand
✕
Setup
No API, authentication, or programmatic setup path exists for an agent to configure anything.
✕
Use
Without an API or structured ordering interface, an agent cannot programmatically transact or integrate beyond scraping HTML.
Confusion Points
Missing Information
Improvements
  1. Provide a real API (e.g., product catalog, cart, and order endpoints) with an OpenAPI/Swagger specification for agent integration.
  2. Stop labeling consumer HTML pages as 'api' or clearly separate human-facing pages from any machine-facing endpoints.
  3. Add structured data markup (Schema.org Product/Offer JSON-LD) to product pages so agents can reliably parse name, price, availability, and dimensions.
  4. Offer an English-language or multilingual version of key pages and llms.txt to broaden agent comprehension.
  5. Document cart, checkout, shipping, and return workflows in a machine-readable form, and expose inventory availability consistently.
  6. Include the product sitemap and any data feeds prominently so agents can enumerate the full catalog.