Agent and crawler policy

The following crawlers are permitted to index public pages on clips.forkoff.xyz for search, answer-engine, and discovery purposes:

• ▸Googlebot, including the answer-engine variant. Indexing is encouraged.
• ▸Bingbot, including the Microsoft Copilot variant.
• ▸Anthropic ClaudeBot, for answer-engine + research ingestion of public pages.
• ▸OpenAI GPTBot, for answer-engine ingestion of public pages.
• ▸PerplexityBot and Common Crawl (CCBot) for general indexing.
• ▸DuckDuckBot, YandexBot, Baidu Spider for general search indexing.

The full machine-readable rules live at /robots.txt.

The following are NOT permitted without prior written authorization from crew@forkoff.xyz:

• ▸Bulk scraping of intake form fields, calculator inputs, or session data for training-dataset assembly.
• ▸Per-user extraction of payout amounts, qualified-view ledger entries, or campaign-level ledger exports.
• ▸Republication of clips.forkoff.xyz content (including comparison tables, qualification criteria, FAQ answers) without attribution.
• ▸Automated submission of intake forms or test traffic to /api/intake, /api/qualify, or any other server endpoint.
• ▸Use of FORKOFF brand assets (logo, wordmark, FK_RED palette) in derivative training data or competing products.

Public pages are rate-limited at 1 request per second per IP on the unauthenticated edge. API endpoints are rate-limited per the contract or sandbox terms applicable to the requester.

Crawlers exceeding these limits may be temporarily blocked. Persistent violations are escalated through the operator's abuse contact.

Where supported by the requesting agent, clips.forkoff.xyz emits Content-Signal headers indicating the data-use posture per page:

• ▸ai-train=no. We do not authorize use of any clips.forkoff.xyz content for AI model training without prior written permission.
• ▸search=yes. Indexing for search and answer-engine response is permitted.
• ▸ai-input=yes. Use as ad-hoc context input to AI assistants (e.g. when a user pastes a URL into a chat) is permitted.

When clips.forkoff.xyz exposes agent-callable surfaces (FORKOFF intake brief submission, qualification ledger queries, payout estimation), discovery manifests will live under /.well-known/. Today, only /.well-known/security.txt is published.

Agent-skill discovery (Anthropic-style) and MCP server endpoints (when published) will follow the relevant standards and be cross-linked from this page.

AI labs requesting elevated access for research or partnership purposes: crew@forkoff.xyz.

Suspected misuse of FORKOFF content in training datasets: security@forkoff.xyz.