# llms.txt — Policy for AI/LLM access and use of content from swaragh.com
# Domain: https://swaragh.com/
# Owner: Swaragh Technologies (Website Design & Development)
# Version: 1.0
# Effective-Date: 2025-08-19
# Contact: info@swaragh.com
#
# Purpose
# -------
# This file communicates Swaragh Technologies' preferences for how artificial
# intelligence systems ("AI", "LLMs", "AI crawlers", "AI assistants") may access,
# cache, and use content from this website. It complements but does not replace
# robots.txt and applicable laws. Vendors should abide by the most restrictive
# of robots.txt and this llms.txt when conflicts arise.
#
# Licensing & Allowed Uses (high level)
# -------------------------------------
# • Reading for real-time assistance ("inference use") is allowed, provided the
#   content is shown to a user on demand with clear source attribution.
# • Training and fine-tuning use is NOT allowed without a separate written license.
# • Indexing for search/snippet generation by general web search is allowed if
#   compliant with robots.txt and attribution rules.
# • Commercial redistribution of our content or embeddings is prohibited without
#   a license.
# • Respect user privacy, applicable laws, and do not collect or store personal
#   data from contact forms, account areas, or non-public endpoints.
#
# Required Behaviors
# ------------------
# • Obey robots.txt, HTTP status codes, and noindex/nofollow/robots meta.
# • Attribute Swaragh Technologies when quoting or summarizing; include a visible
#   link to the exact page URL.
# • Quote Limits: up to 200 words per page without a separate license.
# • Caching: AI systems may cache for up to 30 days for inference. Revalidate on
#   next access or honor ETag/Last-Modified headers. Do not retain training datasets.
# • Rate Limits: Max 1 request/second, 10 requests/minute from a single vendor IP
#   or identified crawler cluster, unless whitelisted.
# • Freshness Embargo: Do not use content published/updated in the last 24 hours
#   for training or model distillation even if training were otherwise licensed.
# • Takedown: Remove specific content from caches or datasets within 10 business
#   days upon request to our Contact.
#
# Prohibited Uses (without explicit written license)
# --------------------------------------------------
# • Model training, fine-tuning, alignment, evaluation datasets, or synthetic data
#   generation derived from our content.
# • Creation of derivative design portfolios/case studies using our content or
#   media (text, images, code) beyond fair use excerpting.
# • Use of private, gated, or non-public content (including but not limited to
#   /wp-admin/, /wp-login.php, form submissions, client portals, draft pages,
#   or materials explicitly marked confidential).
#
# Site Structure: Sitemaps & Discovery
# ------------------------------------
Sitemap: https://swaragh.com/sitemap.xml
# Additional sitemaps (if any):
# Sitemap: https://swaragh.com/post-sitemap.xml
# Sitemap: https://swaragh.com/page-sitemap.xml
# Sitemap: https://swaragh.com/portfolio-sitemap.xml
#
# Path Rules (augmenting robots.txt; AI crawlers MUST treat these as binding)
# ---------------------------------------------------------------------------
# Allowed for inference (read-only access with attribution):
Allow: /
#
# Disallowed for AI access of any kind (including inference, training, or caching):
Disallow: /wp-admin/
Disallow: /wp-login.php
Disallow: /wp-json/wp/v2/users
Disallow: /cart/
Disallow: /checkout/
Disallow: /my-account/
Disallow: /account/
Disallow: /portal/
Disallow: /privacy/*
Disallow: /terms/*
Disallow: /cgi-bin/
Disallow: /?s=*
Disallow: /search/*
Disallow: /wp-content/uploads/private/
#
# Media & Assets
# --------------
# • Low-resolution thumbnails may be displayed in AI results as part of a link
#   preview. Do NOT store or redistribute full-resolution media without a license.
# • Hotlinking is prohibited. Use your own cached copy within the cache window.
#
# Vendor-Specific Directives
# --------------------------
# Unless a custom license is in place, the following vendor policies apply.
# Vendors should map their crawlers/agents to these names and honor them.

[Vendor OpenAI]
Allow-Read: true
Allow-Train: false
Cache-Max-Age: 30d
Attribution-Required: true
Contact: info@swaragh.com

[Vendor Anthropic]
Allow-Read: true
Allow-Train: false
Cache-Max-Age: 30d
Attribution-Required: true

[Vendor Google-Extended]
# Controls use of content for generative AI by Google. Disallowed for training.
Allow-Read: true
Allow-Train: false

[Vendor Applebot-Extended]
Allow-Read: true
Allow-Train: false

[Vendor Perplexity]
Allow-Read: true
Allow-Train: false
Rate-Limit: 10/min

[Vendor CommonCrawl]
Allow-Read: false
Allow-Train: false

[Vendor Amazon]
Allow-Read: true
Allow-Train: false

[Vendor Meta]
Allow-Read: false
Allow-Train: false

[Vendor ByteDance]
Allow-Read: false
Allow-Train: false

[Vendor Baidu]
Allow-Read: false
Allow-Train: false

[Vendor You.com]
Allow-Read: true
Allow-Train: false

# Add additional vendors as needed using the same block format.
#
# HTTP Header Signals (optional but recommended for vendors)
# ----------------------------------------------------------
# Swaragh may emit the following headers server-side to reinforce intent:
#   X-AI-Use: inference-only
#   X-AI-Train: disallow
#   Cache-Control: public, max-age=86400, stale-while-revalidate=86400
# Vendors should prioritize llms.txt over speculative heuristics.
#
# Citation & Excerpt Rules
# ------------------------
# • Include the page title and a canonical URL.
# • Total excerpt length per response must not exceed 200 words without license.
# • Summaries must be truthful and not suggest endorsement.
#
# Legal
# -----
# • Copyright © Swaragh Technologies. All rights reserved.
# • No assignment of rights is granted via this file. This is a permission notice.
# • Any conflicts are governed by Indian law and the courts of Bangalore, India.
#
# Takedown Procedure
# ------------------
# Email our Contact with: (1) exact URL(s), (2) nature of use, (3) vendor name,
# (4) crawler IPs, (5) desired remedy. We will respond within 10 business days.
#
# Machine-Readable Policy (JSON)
# ------------------------------
# Vendors may parse the block below.

{ "version": "1.0",
  "owner": "Swaragh Technologies",
  "domain": "swaragh.com",
  "effective_date": "2025-08-18",
  "policy": {
    "allow_read": true,
    "allow_train": false,
    "cache_max_age_days": 30,
    "quote_limit_words": 200,
    "attribution_required": true,
    "rate_limit": { "rps": 1, "rpm": 10 },
    "embargo_hours_for_training": 24
  },
  "paths": {
    "allow": ["/"],
    "disallow": [
      "/wp-admin/","/wp-login.php","/wp-json/wp/v2/users",
      "/cart/","/checkout/","/my-account/","/account/","/portal/",
      "/privacy/*","/terms/*","/cgi-bin/","/?s=*","/search/*",
      "/wp-content/uploads/private/"
    ]
  },
  "sitemaps": [
    "https://swaragh.com/sitemap.xml"
  ],
  "vendors": {
    "openai": {"allow_read": true, "allow_train": false},
    "anthropic": {"allow_read": true, "allow_train": false},
    "google_extended": {"allow_read": true, "allow_train": false},
    "applebot_extended": {"allow_read": true, "allow_train": false},
    "perplexity": {"allow_read": true, "allow_train": false},
    "commoncrawl": {"allow_read": false, "allow_train": false},
    "amazon": {"allow_read": true, "allow_train": false},
    "meta": {"allow_read": false, "allow_train": false},
    "bytedance": {"allow_read": false, "allow_train": false},
    "baidu": {"allow_read": false, "allow_train": false},
    "youdotcom": {"allow_read": true, "allow_train": false}
  },
  "takedown_contact": "info@swaragh.com"
}

# End of llms.txt