Trusted Collaboration Tunnel (TCT)
Efficient, verifiable content delivery for AI crawlers. Reduce bandwidth by 83%, tokens by 86%.
The Trusted Collaboration Tunnel protocol enables efficient machine-readable content delivery through bidirectional discovery, template-invariant fingerprinting, and sitemap-first optimization. Patent-pending (US 63/895,763).
Measured Results from 970+ Production URLs
83% Bandwidth Savings
103 KB → 17.7 KB average per fetch
86% Token Reduction
13,900 → 1,960 tokens for AI processing
90%+ Skip Rate
Unchanged content detected via ETag
970+ URLs
Across 3 production sites with 100% compliance
8,158 GWh/year
Potential energy savings at 10% adoption
How It Works: 4-Part Protocol
1. Bidirectional Discovery
- C-URL contains:
<link rel="alternate" type="application/json" href="M-URL"> - M-URL responds:
Link: <C-URL>; rel="canonical" - Verifiable handshake prevents URL guessing
2. Template-Invariant Fingerprinting
- Content normalized (strip HTML, CSS, JS)
- SHA-256 hash → ETag header
- Same content = same hash, regardless of theme changes
3. Conditional Request Discipline
- Client sends:
If-None-Match: "sha256-abc123..." - Server responds:
304 Not Modified(zero body) - 90%+ of requests skip full fetch
4. Sitemap-First Verification
/llm-sitemap.jsonlists all URLs + hashes- Client compares local cache vs sitemap
- Only fetch changed content
- Zero-fetch optimization
Production Sites (100% Compliant)
- wellbeing-support.com (400 URLs, 10/10)
- galaxybilliard.club (500 URLs, 10/10)
- bestdemotivationalposters.com (500 URLs, 9/10)
Test Your Implementation: llmpages.org/validator
Patent & Licensing
Patent Status: US 63/895,763 (Filed October 8, 2025) – Patent Pending
Free for Website Owners: GPL v2+ WordPress plugin, MIT Cloudflare Worker, no restrictions for your own sites.
Commercial Licensing: AI companies, CDN providers, crawler operators at scale (>10K URLs/month) should contact [email protected] for licensing terms.
IETF Standards Track
Target: IETF 125 (March 2026) | Working Group: GREEN (energy efficiency)
Network savings: 0.06 kWh/GB reduced | AI inference savings: 0.000187 kWh/token reduced
