AEO Agent — AI Search Visibility Monitor

Autonomous agent that measures and improves your brand's visibility in AI-powered search engines (ChatGPT, Perplexity, Claude, Gemini, Google AI Overviews).

What is AEO?

Answer Engine Optimization is the practice of optimizing web content to be discovered, understood, and cited by AI search systems. Unlike traditional SEO that targets PageRank and SERP positions, AEO focuses on:

• Entity recognition — AI models correctly identifying your brand
• Citation probability — AI models citing your website as a source
• Answer inclusion — appearing in AI-generated answers to relevant queries
• Knowledge accuracy — AI models having correct information about you

Architecture

18 tools across 5 categories:

VISIBILITY (4)          ENTITY (3)              CONTENT (4)
├── Multi-model test    ├── Entity recognition  ├── Content AI analysis
├── Citation tracking   ├── Knowledge graph     ├── Schema audit
├── AI Overview check   └── Entity relationships├── llms.txt optimization
└── Competitor benchmark                        └── FAQ generation

STRATEGY (3)            ACTION (4)
├── AEO improvements    ├── GitHub PR creation
├── Citation opps       ├── HTML report generation
└── Query patterns      ├── Discord notifications
                        └── Structured data generation

Configuration

Copy .env.example to .env.local and fill in your values:

cp .env.example .env.local

Required Environment Variables

Variable	Description	Example
ANTHROPIC_API_KEY	Anthropic API key (agent model)	sk-ant-...
OPENAI_API_KEY	OpenAI API key (multi-model testing)	sk-...
GOOGLE_GENERATIVE_AI_API_KEY	Google AI key (multi-model testing)	AI...
SITE_URL	Your website URL	https://mystore.com
COMPANY_NAME	Your brand name	Acme Electronics

Recommended Environment Variables

Variable	Description	Example
INDUSTRY_CONTEXT	Your industry (used in AI prompts)	e-commerce electronics
TARGET_KEYWORDS	Comma-separated keywords to monitor	best laptops, gaming PC
COMPETITORS	Comma-separated competitor domains	bestbuy.com, newegg.com
GITHUB_REPO	GitHub repo for PR creation (owner/repo)	acme/website
GITHUB_TOKEN	GitHub PAT with repo scope	ghp_...
DISCORD_WEBHOOK_URL	Discord webhook for notifications	https://discord.com/api/webhooks/...
GSC_*	Google Search Console OAuth credentials	See .env.example
BLOB_READ_WRITE_TOKEN	Vercel Blob for persistent storage	vercel_blob_...

Configuration Examples

E-commerce store:

COMPANY_NAME=TechShop
SITE_URL=https://techshop.com
INDUSTRY_CONTEXT=e-commerce consumer electronics
TARGET_KEYWORDS=best laptops 2026, gaming monitors, wireless headphones
COMPETITORS=bestbuy.com, newegg.com, amazon.com

SaaS company:

COMPANY_NAME=DataFlow
SITE_URL=https://dataflow.io
INDUSTRY_CONTEXT=B2B SaaS data analytics platform
TARGET_KEYWORDS=data analytics platform, business intelligence tool
COMPETITORS=tableau.com, looker.com, metabase.com

Stack

• Runtime: Next.js (App Router) on Vercel
• Agent: Vercel AI SDK v6 (ToolLoopAgent)
• Model: Claude Sonnet 4 via @ai-sdk/anthropic
• Multi-model testing: @ai-sdk/openai (GPT), @ai-sdk/google (Gemini)
• Notifications: Discord webhooks
• Code changes: GitHub PRs via REST API
• Data: Google Search Console (OAuth), Google Analytics 4

Tools

Visibility

Tool	Description
testMultiModelVisibility	Query Claude/GPT/Gemini with industry questions, analyze mentions and citations
trackCitationSources	Historical analysis of cited sources, trends, brand mention rate
analyzeAIOverviewPresence	Check Google AI Overviews for brand presence
benchmarkCompetitorsAEO	Compare visibility against competitors

Entity & Knowledge

Tool	Description
analyzeEntityRecognition	Test if AI models know your brand (identity, pricing, industry)
auditKnowledgeGraph	Check Wikidata, Google KP, Crunchbase, Wikipedia presence
mapEntityRelationships	Map brand connections to industry topics

Content Optimization

Tool	Description
analyzeContentForAI	Score page content for AI citability (8 criteria, 0-100)
auditSchemaForAI	Check JSON-LD for AI-relevant types (FAQPage, Speakable, HowTo)
optimizeLlmsTxt	Audit and optimize llms.txt for AI crawlers
generateFAQContent	Generate FAQ pairs from real AI model questions with schema

Strategy

Tool	Description
suggestAEOImprovements	Prioritized improvements based on all collected data
identifyCitationOpportunities	Find sources AI models trust, map citation paths
analyzeQueryPatterns	Discover questions people ask AI about your industry

Action

Tool	Description
createGitHubPR	Create PRs with schema, FAQ, llms.txt fixes
writeAEOReport	Generate HTML reports with scores and trend charts
notifyTeam	Discord notifications with structured action items
generateStructuredData	Generate JSON-LD (Organization, SoftwareApp, FAQ, HowTo, Speakable)

Cron Schedule

Schedule	Job	What it does
Weekly	Weekly Visibility	Multi-model test, citation tracking, competitor benchmark, Discord report
Monthly	Monthly Deep Analysis	Full audit (entity, schema, content, knowledge graph), auto-fix PR, HTML report

Quick Start

npm install
cp .env.example .env.local
# Fill in your API keys and site configuration
npm run dev

Open http://localhost:3000 and use the quick actions or type a custom prompt.

How It Works

1. Weekly pulse — Tests AI model visibility, tracks citations, benchmarks competitors. Reports via Discord.
2. Monthly deep dive — Full audit of entity recognition, knowledge bases, content structure, schema markup. Auto-fixes what it can and creates a single PR. Generates HTML report. Escalates human-required actions with clear instructions.
3. Interactive chat — Use the web UI for ad-hoc queries and analysis.

License

MIT