# =============================================================================
# Carol Whaley Addassi, Fine Art Photography
# https://caroladdassi.com
#
# This site is intentionally discoverable by search engines and by AI assistants
# that *answer questions and link back* to the original source.
#
# This site is NOT licensed for any form of AI / ML training, dataset inclusion,
# fine-tuning, or generative output. Crawlers operated for the purpose of
# building training datasets, or for ingesting content into generative models
# that produce derivative imagery, are not permitted.
#
# This file uses both the long-established User-agent / Allow / Disallow
# directives AND the newer Content-Signals declaration that Cloudflare and
# others enforce in 2025+.
# =============================================================================

# -----------------------------------------------------------------------------
# Content-Signals policy (IETF-style declaration)
# -----------------------------------------------------------------------------
# search:   YES   building a search index, returning hyperlinks + short excerpts
# ai-input: NO    retrieval-augmented generation, real-time grounding for LLMs,
#                 AI-generated search summaries that use this content
# ai-train: NO    training or fine-tuning of AI / ML models on this content
# -----------------------------------------------------------------------------
Content-Signal: search=yes, ai-input=no, ai-train=no


# =============================================================================
# SEARCH and DISCOVERY agents, explicitly ALLOWED
# These crawlers index the site and return links / short excerpts so that
# collectors, designers, journalists, and curators can find the artist.
# =============================================================================

User-agent: Googlebot
Allow: /

User-agent: Googlebot-Image
Allow: /

User-agent: Bingbot
Allow: /

User-agent: DuckDuckBot
Allow: /

User-agent: Applebot
Allow: /

# OpenAI's search index (NOT the training crawler). This bot powers
# "search the web" inside ChatGPT and returns links back to this site.
User-agent: OAI-SearchBot
Allow: /

# When a ChatGPT user explicitly asks ChatGPT to read a URL.
User-agent: ChatGPT-User
Allow: /

# When a Perplexity user explicitly asks Perplexity to read a URL.
User-agent: Perplexity-User
Allow: /

# When a Claude user explicitly asks Claude to read a URL.
User-agent: Claude-User
Allow: /

# Anthropic's question-answering search bot (not the training crawler).
User-agent: Claude-SearchBot
Allow: /

# Mojeek, Kagi, You.com, etc.
User-agent: MojeekBot
Allow: /

User-agent: KagiBot
Allow: /

User-agent: YouBot
Allow: /


# =============================================================================
# TRAINING / DATASET agents, explicitly DISALLOWED
# These crawlers exist to ingest content into AI training datasets or into
# generative-AI pipelines. Carol Whaley Addassi does not license her work
# for any such use.
# =============================================================================

# Common Crawl, the dataset that feeds nearly every open-source image
# and language model. This is the single most important block.
User-agent: CCBot
Disallow: /

# OpenAI training crawler (separate from OAI-SearchBot above).
User-agent: GPTBot
Disallow: /

# Google's AI / Gemini training crawler (separate from Googlebot above).
User-agent: Google-Extended
Disallow: /

# Anthropic training crawlers (separate from Claude-SearchBot above).
User-agent: anthropic-ai
Disallow: /

User-agent: ClaudeBot
Disallow: /

User-agent: Claude-Web
Disallow: /

# Apple Intelligence training crawler (separate from Applebot above).
User-agent: Applebot-Extended
Disallow: /

# Meta / Facebook AI ingestion.
User-agent: Meta-ExternalAgent
Disallow: /

User-agent: Meta-ExternalFetcher
Disallow: /

User-agent: FacebookBot
Disallow: /

# ByteDance / TikTok ingestion (widely documented training use).
User-agent: Bytespider
Disallow: /

# Amazon training crawler.
User-agent: Amazonbot
Disallow: /

# Perplexity's general crawler. Perplexity-User above is permitted for
# user-initiated reads, but the broad PerplexityBot has documented
# training behavior and is disallowed.
User-agent: PerplexityBot
Disallow: /

# Cohere training and dataset crawlers.
User-agent: cohere-ai
Disallow: /

User-agent: cohere-training-data-crawler
Disallow: /

# Diffbot ingests structured content for AI products.
User-agent: Diffbot
Disallow: /

# Omgili / webz.io, sold into AI dataset pipelines.
User-agent: Omgili
Disallow: /

User-agent: Omgilibot
Disallow: /

# AI2 / Allen Institute scraping for OLMo and other model training.
User-agent: AI2Bot
Disallow: /

User-agent: ai2bot-dolma
Disallow: /

# Image-specific scrapers used to build text-to-image datasets.
User-agent: ImagesiftBot
Disallow: /

User-agent: img2dataset
Disallow: /

# Hugging Face dataset crawlers.
User-agent: huggingface-datasets
Disallow: /

# Timpi
User-agent: TimpiBot
Disallow: /


# =============================================================================
# Default for any other agent
# =============================================================================
# All other user agents (search and otherwise) may index the site, but the
# Content-Signal block above declares that AI training and AI input are NOT
# permitted under any reading of this site's terms.
User-agent: *
Allow: /
Disallow: /admin
Disallow: /admin/

Sitemap: https://caroladdassi.com/sitemap.xml