AI Rank Optimizer (LLM + AGENT Frameworks)

Rewrite or proofread content so it is easy for:

1. LLM answer engines (ChatGPT, Claude, Perplexity) to extract, cite, and recommend
2. Autonomous AI agents to parse, compare, and take action

Two Audiences, Different Needs

LLM Answer Engines (Human-in-the-loop)

• Humans ask questions, AI provides answers citing your content
• Optimize for: extraction, citation, featured snippets
• Key: Answer-first, quotable content, trust signals

Autonomous AI Agents (No human)

• Agents browse, compare products, make decisions, execute tasks
• Optimize for: machine parsing, structured data, actionable endpoints
• Key: Structured facts, API access, decision-ready data

LLM Framework (for Answer Engines)

1. Answer-first

• First sentence gives the direct answer
• Include: definition + who it's for + when it's NOT for
• Write quotable 2-5 sentence summaries LLMs will extract

2. Intent-matched headings

• H2/H3 titles match target queries verbatim
• Correct heading hierarchy (no skipping levels)
• Use question-format headings for FAQ sections

3. Clear structure

• Lists, tables, steps, FAQs over paragraphs
• "Quick summary" and "Key takeaways" sections
• Comparison tables for vs-queries

4. Schema markup

• FAQPage JSON-LD for Q&A content
• HowTo schema for tutorials/guides
• Product/SoftwareApplication for products
• Article schema with author/date for blog posts

5. Trusted sources

• Cite quantitative claims or label "internal data"
• Link to primary sources
• Include dates, version numbers, last-updated timestamps

6. Unique perspective

• Unique frameworks, decision trees, checklists
• Original benchmarks, comparisons, data
• Proprietary methodology or rubrics

AGENT Framework (for Autonomous AI Agents)

A - Accessible structured data

• JSON-LD schema on every page
• Machine-readable pricing tables (not images)
• Structured product specs, limits, requirements
• RSS/Atom feeds for updates
• Comprehensive sitemap.xml

G - Grounded facts for decisions

• Explicit pricing with currency and billing cycle
• Hard limits and quotas (not "unlimited*")
• Compatibility matrices (platforms, versions, integrations)
• SLAs, uptime guarantees, support tiers
• Comparison tables vs alternatives

E - Endpoints for action

• API documentation with examples
• Webhook/integration setup guides
• Direct links to signup, trial, purchase
• Contact/support endpoints
• Status page URLs

N - Navigable hierarchy

• Consistent URL structure
• Breadcrumbs in markup
• Clear content taxonomy
• robots.txt allowing AI crawlers
• Discovery files (llms.txt, agents.txt)

T - Trust markers for machines

• Security certifications (SOC2, GDPR, etc.)
• Published changelog/release notes
• Public roadmap or feature status
• Customer logos/testimonials with verifiable details
• Third-party reviews with links

Discovery Standards & Protocols

For LLM Answer Engines

llms.txt

• Location: /llms.txt
• Purpose: Help LLMs understand site content at inference time
• Format: Markdown with H1 (site name), summary, key page links
• Spec: https://llmstxt.org/

# Your Company Name

> One-line description of what you do.

## Docs
- [Getting Started](https://docs.example.com/start)
- [API Reference](https://docs.example.com/api)

## Products
- [Product Name](https://example.com/product): Description

## Pricing
- [Pricing](https://example.com/pricing)

For Autonomous Agents

agents.txt

• Location: /agents.txt
• Purpose: B2A (Business to Agent) service discovery
• Spec: https://agentstxt.dev/

# agents.txt
name: Your Company
description: What agents can do with your service
api_endpoint: https://api.example.com
auth_type: api_key
documentation: https://docs.example.com/api
mcp_server: https://mcp.example.com
capabilities:
  - read_data
  - write_data
  - transactions

/.well-known/api-catalog (RFC 9727)

• Location: /.well-known/api-catalog
• Purpose: API endpoint discovery (like robots.txt for APIs)
• Format: JSON

{
  "apis": [
    {
      "name": "Your API",
      "description": "What it does",
      "url": "https://api.example.com",
      "documentation": "https://docs.example.com/api",
      "type": "REST"
    }
  ]
}

Agent Card (A2A Protocol)

• Purpose: Agent-to-Agent protocol capabilities
• Format: JSON endpoint
• Spec: https://a2a-protocol.org/

{
  "name": "Your Agent",
  "description": "What this agent does",
  "capabilities": ["task1", "task2"],
  "endpoint": "https://agent.example.com",
  "auth": {"type": "oauth2"}
}

MCP Registry

• Purpose: Make your MCP server discoverable
• Registry: https://registry.modelcontextprotocol.io/
• Submit: https://github.com/modelcontextprotocol/registry

Master Checklists

Page-Level Checklist

Run this for every content page:

Landing Pages

• First sentence directly answers "what is this?"
• H1 matches primary target query
• H2s match secondary target queries
• Comparison table vs competitors (if applicable)
• "Who it's for" and "Who it's NOT for" sections
• Pricing in a table with explicit numbers
• Hard limits/quotas stated (not "unlimited*")
• FAQ section with question-format headings
• Customer quote with attribution
• Clear CTA with direct signup link
• FAQPage JSON-LD schema
• Product/SoftwareApplication JSON-LD schema
• Last-updated date visible

Documentation Pages

• Problem/solution stated in first paragraph
• Prerequisites listed upfront
• Step-by-step instructions (numbered)
• Code examples with language tags
• Expected output shown
• Common errors and solutions
• Links to related docs
• HowTo JSON-LD schema
• Last-updated date visible

Blog Posts

• Answer/thesis in first paragraph
• Key takeaways section (top or bottom)
• Specific numbers and data points
• Citations for external claims (links)
• "Internal data" label for proprietary stats
• Author name and date
• Internal links to product/docs
• Article JSON-LD schema

Pricing Pages

• All plans in a comparison table
• Explicit prices with currency
• Billing cycle stated (monthly/annual)
• Feature limits per plan (numbers, not checkmarks)
• API rate limits
• Support response times
• SLA/uptime guarantee
• Enterprise contact method
• Free trial/plan details
• Product JSON-LD with Offer schema

API/Developer Docs

• Authentication methods documented
• Base URL clearly stated
• All endpoints listed with methods
• Request/response examples
• Rate limits documented
• Error codes explained
• SDK/client library links
• Webhook payload examples
• Changelog/versioning info

Integration Pages

• All integrations in tables (not prose)
• Status per integration (native/plugin/beta)
• Setup method per integration
• Data sync frequency
• What data is sent/received
• Link to setup docs per integration

Site-Level Checklist

Run this once for the entire site:

Discovery Files (Create These)

• /robots.txt - allows AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Googlebot)
• /llms.txt - site summary for LLM inference
• /agents.txt - agent service discovery
• /.well-known/api-catalog - API discovery (RFC 9727)
• /sitemap.xml - comprehensive, up-to-date
• /feed.xml or /rss.xml - blog/changelog feed

robots.txt Template

User-agent: GPTBot
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Google-Extended
Allow: /

User-agent: *
Allow: /

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

AI Crawler Access

• GPTBot allowed (OpenAI)
• ClaudeBot allowed (Anthropic)
• PerplexityBot allowed
• Goog