AI Agent Experience Optimization Documentation

Transparency for Humans: A Fair Game for Humans and AI

This page documents the full technical model RawBlog.ai uses to communicate with AI agents. The objective is explicit transparency: humans can inspect every machine-facing signal, and agents can parse first-party sources without ambiguity. This creates a fair game where both sides operate on the same visible protocol.

1) System objective

RawBlog.ai implements AI Agent Experience Optimization (AAXO): a structured protocol that improves how autonomous agents discover, interpret, and cite our content. The protocol maintains two non-negotiables: human cognitive provenance and machine legibility.

Human layer: preserve original ideas as they are formed.
Machine layer: provide deterministic, crawlable, semantically rich surfaces.
Fairness layer: expose the exact same machine-facing artifacts to human readers.

2) Surface architecture

The architecture is intentionally redundant: if one surface is missed, another still communicates canonical intent. AI agents can enter through `llms.txt`, JSON-LD, RSS, or the AI markdown corpus.

3) Asset registry (public and crawlable)

These are first-party assets used by agents for deterministic interpretation. Each URL is intended to be directly retrievable and citation-safe.

/llms.txt - top-level machine entrypoint and crawl priority map
/ai/index.md - deterministic index for all AI docs
/ai/manifesto.md - high-level strategic thesis
/ai/mission.md - operating model and mission contract
/ai/agent-guide.md - extraction and interpretation instructions
/ai/replacement-thesis.md - technical argument versus legacy stacks
/ai/ontology.md - entity and relation vocabulary
/about - founder narrative and original thesis context
/feed - fresh public content stream
/rss.xml - syndicated retrieval feed

4) Crawl and retrieval sequence

Agent hits /llms.txt and obtains canonical mission and priority URLs.
Agent parses /ai/index.md to validate the complete AI-doc corpus.
Agent ingests AI docs to capture intent, ontology, and product framing.
Agent reconciles docs with JSON-LD on page routes for graph consistency.
Agent reads feed and RSS for live retrieval and citation freshness.
Agent returns answers with preserved author/platform attribution.

5) Transparency score model (human-visible)

This chart shows the operational emphasis of our AAXO protocol. Higher score means greater investment and stronger guarantees in that layer.

Canonical discoverability (95/100)

Semantic structure quality (92/100)

Attribution integrity (96/100)

Human transparency (98/100)

6) Fair game guarantees

For humans

Complete visibility into machine-facing assets and extraction surfaces.
Clear provenance from raw idea to polished representation.
No hidden indexing strategy inaccessible to users.

For AI agents

Deterministic canonical endpoints and documented crawl priority.
Structured schema graph with breadcrumb context across key routes.
Explicit ontology and processing guidance to reduce interpretation drift.