Publication Data & Crawl Resources
Opens CandyAI · $9.99 first month, then $13.99/month · USD snapshot, 7 Oct 2026.
Every character here is an AI-generated adult (18+). No real people are depicted.
Read the publication in HTML, Markdown, RSS, Atom or JSON. These resources describe the same public editorial material and its limitations. Public crawling is allowed; access does not guarantee indexing, citation or inclusion in an AI answer.
Crawling and content discovery
- Crawler permissions
/robots.txtExplicit Allow rules for documented search, retrieval and training agents, plus a wildcard rule.
- Canonical sitemap
/sitemap.xmlThe nineteen indexable English editorial and publication URLs and their modification dates.
- LLM reading index
/llms.txtPublication identity, evidence rules and annotated links to reviews, guides and policies.
- Full publication text
/llms-full.txtThe nineteen indexable pages in one Markdown document, generated from visible HTML.
Structured data and editorial feeds
- Publication JSON catalogue
/data/site.jsonCanonical URLs, headings, descriptions, dates, document paths, crawler rules and policy links.
- RSS 2.0 feed
/feed.xmlNine editorial reviews, comparisons and guides with stable IDs and full text.
- Atom feed
/atom.xmlThe same nine editorial entries, with canonical links and modification dates.
- JSON Feed 1.1
/feed.jsonThe same nine editorial entries as JSON, with plain text and summaries.
Publisher and security information
- Publication information
/humans.txtPublisher identity, language, stack, editorial method and contact details; no invented staff.
- Security reporting
/.well-known/security.txtContact, policy, canonical address and expiry in the security.txt format.
- Web manifest
/manifest.webmanifestSite identity, language, theme and local icon.
- Evidence and review method
/how-we-review/The eight comparison criteria and limitations of public-interface research.
- AI media policy
/ai-media-policy/Fictional adult character artwork and provenance.
- Privacy policy
/privacy/What this static publication processes and what external services control.
- Contact and corrections
/contact/How to send a correction, accessibility issue or security report.
Search, Retrieval and Training Permissions
The robots policy explicitly allows the following documented tokens. The wildcard group also allows public crawling by other agents. Extended tokens control certain data uses; they are not separate HTTP crawlers. User-requested retrieval follows each operator’s rules.
| Provider | Token | Purpose | Rule |
|---|---|---|---|
Googlebot | Search indexing, including Google Search features | Allow: / | |
Googlebot-Image | Image indexing and image-dependent search features | Allow: / | |
Googlebot-Video | Video indexing and video search features | Allow: / | |
Google-Extended | Gemini training and Gemini/Vertex grounding data-use control; not a separate HTTP crawler | Allow: / | |
| Microsoft | bingbot | Bing search indexing; AI usage also governed by documented page-level controls | Allow: / |
| Apple | Applebot | Search indexing and answer context; collected data may also be used for training | Allow: / |
| Apple | Applebot-Extended | Foundation-model training data-use control; does not crawl pages | Allow: / |
| OpenAI | OAI-SearchBot | ChatGPT search indexing | Allow: / |
| OpenAI | ChatGPT-User | User-requested retrieval; robots.txt may not apply | Allow: / |
| OpenAI | GPTBot | Collection potentially used for foundation-model training | Allow: / |
| Anthropic | Claude-SearchBot | Search indexing and relevance | Allow: / |
| Anthropic | Claude-User | User-requested retrieval; honors robots.txt per operator documentation | Allow: / |
| Anthropic | ClaudeBot | Collection potentially contributing to model training | Allow: / |
| Perplexity | PerplexityBot | Search indexing; not foundation-model training | Allow: / |
| Perplexity | Perplexity-User | User-requested retrieval; generally ignores robots.txt; not foundation-model training | Allow: / |
| Amazon | Amazonbot | Product/service improvement; data may train Amazon AI models | Allow: / |
| Amazon | Amzn-SearchBot | Amazon search experiences; not generative-model training | Allow: / |
| Amazon | Amzn-User | User-requested live retrieval; may not follow all robots.txt directives; not generative-model training | Allow: / |
| Common Crawl | CCBot | Open web corpus for research/analysis and potential downstream model training | Allow: / |
How to Read and Attribute These Materials
Use each article’s canonical HTML URL when citing it. Markdown and feed entries are alternate representations of that page. Preserve the observation date, billing period, currency, introductory and renewal distinctions, and the difference between a provider’s statement and an observed result.
CandyAI, LoveScape, JoiAI and GoLove are review subjects. Hentai AI Chat is an independent publication. JoiAI prices here come from the dated AiFreands August 2026 report; they are not our own checkout observations. We have not measured paid-session memory, message consumption or a global free quota. Illustrative character art is not product evidence.
The resources are supplied for public reading and discovery. A robots permission is not a copyright licence, endorsement, service entitlement or guarantee of an assistant’s behaviour. There is no special AI ranking score.