Package Exports

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@aichatwatch/aicw) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

AI Chat Watch (AICW) – Monitor AI Engines & Maximize Your AI Visibility

AICW is a comprehensive monitoring and optimization suite designed to help you dominate the emerging AI search landscape. Track how AI engines like ChatGPT, Perplexity, Claude, and others discover, cite, and recommend your brand—while giving you actionable intelligence to improve your AI visibility.

What You Get:

AI Search Results Tracking:

Track brand, product, and competitor mentions across major AI search engines
Track how often mentioned, position in search results, share of voice
Review sources for citations for specific items
Review sources for type of content, domains, and website types
Analyze trends over time

AI Visibility Auditing:

Assess how "AI-readable" your website and content are
Identify technical barriers preventing AI engines from discovering your content
Get actionable recommendations to optimize for AI and LLMs
Validate that your structured data, metadata, and content architecture are AI-friendly

AICW is free, open-source and runs on your computer, you own all your data.

Explore sample reports: https://aichatwatch.com/demo/reports/

ELI5 (Explain like I'm 5 years old)?: AI Chat Watch is like a notebook that keeps track of which toys or pizza places the robot talks about and if robots can see them or not!

🚀 Quick Start

Grade AI visibility of your website

Run Terminal app and paste this command:

npx @aichatwatch/aicw ai-visibility YOURWEBSITE.com

Run in interactive mode to create and run detailed AI visibility report

npx @aichatwatch/aicw

NOTE if you are getting errors about access denied or permissions, then you may need to add sudo prefix to the command to run it as administrator, for example:

sudo npx @aichatwatch/aicw ai-visibility YOURWEBSITE.com

Installation (Recommended for continuous monitoring)

Run Terminal app and paste this command:

npm install -g @aichatwatch/aicw

After installation, run aicw just by typing aicw and hitting Enter. It will start AICW in interactive mode.

aicw

Requirements:

Node.js 18+ (Download)
API key from OpenRouter (recommended, free tier available) or OpenAI or any other AI model provider.

Cost: The software is free. You only pay for AI usage through OpenRouter/OpenAI/Perplexity as you consume it:

Typical cost: $0.01-$0.10 per answer + $0.01-$0.05 per question for enrichment/analysis
Example: 5 questions × 3 AI models = 15 answers + enrichment ≈ $0.30-$1.75 total (depending on which models you select)
You control costs by choosing which models to query (free, mid-tier, or premium) and how many questions to ask.

🎬 AICW in Action

AI Visibility Report

AICW AI Visibility Report

📊 All Sample Reports

https://aichatwatch.com/demo/reports/

Sample reports generated by AICW:

Best Cities for Nomads - Travel & lifestyle tracking
Top AI Tools - Technology & platforms
Y Combinator - Analyzing specific topic (Y Combinator)
High Blood Pressure & Hypertension - Healthcare & treatment recommendations from AI models
Best Lawyer for a Startup - Legal services & professionals
and other sample reports

→ Explore all at https://aichatwatch.com/demo/reports/

💾 Data Storage

AICW stores all data locally in platform-specific directories:

macOS: ~/Library/Application Support/aicw/default-user/data/
Windows: %APPDATA%\aicw\default-user\data\
Linux: ~/.config/aicw/default-user/data/

Data Privacy: All data stays on your machine. AICW doesn't send anything to third parties (except API calls to AI providers you configure).

🛠️ Development & Contributing

AICW is built with TypeScript and designed for extensibility.

Local Development

git clone https://github.com/aichatwatch/aicw.git
cd aicw
npm install
npm run build
npm link

# Now 'aicw' command uses your local version
aicw

We welcome contributions! See CONTRIBUTING.md for guidelines.

📊 Pipeline Architecture

Last updated: 2025-10-14

AICW uses a multi-stage pipeline to transform AI responses into actionable insights. Here's the complete build pipeline flow:

flowchart TD
    Start([Start]) --> Setup{Select Pipeline}

    Setup -->|New Project| NewProject[Project: Create New]
    Setup -->|Build| PrepFolders[Prepare Folders]
    Setup -->|Rebuild| PrepFolders

    NewProject --> PrepFolders

    subgraph Setup["Setup Phase"]
        PrepFolders[Prepare Question Folders]
        CleanupCompiled[Cleanup: Remove Old Data]
        CleanupOrphaned[Cleanup: Remove Orphaned Questions]
        DataFileCreate[Create Data Files]
    end

    PrepFolders --> CleanupCompiled
    CleanupCompiled --> CleanupOrphaned
    CleanupOrphaned --> DataFileCreate

    DataFileCreate --> CheckPipeline{Which Pipeline?}
    CheckPipeline -->|Full Build| FetchAnswers
    CheckPipeline -->|Rebuild| ExtractPrompt

    subgraph DataCollection["Data Collection"]
        FetchAnswers[Fetch Answers from AI Models]
    end

    FetchAnswers --> ExtractPrompt

    subgraph EntityExtraction["🔍 Entity Extraction"]
        ExtractPrompt[Prepare Entity Extraction Prompts]
        ExtractEntities[Extract Entities using AI]
        ExtractLinks[Extract Links from Answers]
    end

    ExtractPrompt --> ExtractEntities
    ExtractEntities --> ExtractLinks

    subgraph LinkProcessing["🔗 Link Processing"]
        GetLinkType[Classify Links by Pattern]
        GetLinkTypeAI[AI Classification for Unclassified Links]
        GenLinkTypes[Generate LinkTypes Array]
        GenLinkDomains[Generate LinkDomains Array]
    end

    ExtractLinks --> GetLinkType
    GetLinkType --> GetLinkTypeAI
    GetLinkTypeAI --> GenLinkTypes
    GenLinkTypes --> GenLinkDomains

    subgraph Enrichment["Enrichment & Calculations"]
        CalcMentions[Calculate Mentions]
        CalcAppearance[Calculate Appearance Order]
        CalcInfluence[Calculate Influence Scores]
        CalcTrends[Calculate Historical Trends]

        LinkTypesMentions[Calculate LinkTypes Mentions]
        LinkTypesAppearance[Calculate LinkTypes Appearance]
        LinkTypesInfluence[Calculate LinkTypes Influence]
        LinkTypesTrends[Calculate LinkTypes Trends]

        GenLinks[Generate Entity URLs using AI]
        GenSimilar[Generate Similar Terms using AI]
        GenSummary[Generate AI Summary]
    end

    GenLinkDomains --> CalcMentions
    CalcMentions --> CalcAppearance
    CalcAppearance --> CalcInfluence
    CalcInfluence --> CalcTrends

    CalcTrends --> LinkTypesMentions
    LinkTypesMentions --> LinkTypesAppearance
    LinkTypesAppearance --> LinkTypesInfluence
    LinkTypesInfluence --> LinkTypesTrends

    LinkTypesTrends --> GenLinks
    GenLinks --> GenSimilar
    GenSimilar --> GenSummary

    subgraph ReportGeneration["📊 Report Generation"]
        CleanupReports[Cleanup: Remove Old Reports]
        GenAnswersFile[Generate Answers File]
        GenReport[Generate HTML Reports]
        GenNavigation[Generate Project Navigation]
        ShowSuccess[Show Success Message]
    end

    GenSummary --> CleanupReports
    CleanupReports --> GenAnswersFile
    GenAnswersFile --> GenReport
    GenReport --> GenNavigation
    GenNavigation --> ShowSuccess

    ShowSuccess --> End([End])

    style Start fill:#90EE90
    style End fill:#90EE90
    style Setup fill:#87CEEB
    style DataCollection fill:#98FB98
    style EntityExtraction fill:#FFD700
    style LinkProcessing fill:#FFA500
    style Enrichment fill:#DDA0DD
    style ReportGeneration fill:#87CEFA
    style NewProject fill:#FFB6C1

Pipeline Types

pipeline-project-new: Create a new project with questions → Auto-runs pipeline-project-build
pipeline-project-build: Complete pipeline - Fetch fresh AI answers, extract entities, enrich data, generate reports
pipeline-project-rebuild: Rebuild from existing answers - Skip fetch step, reprocess everything else
pipeline-project-rebuild-report-only: Generate reports only - Use existing enriched data, regenerate HTML

Key Stages

Setup Phase (cyan): Prepares project structure and cleans old data
Data Collection (green): Queries AI models for answers (skipped in rebuild)
Entity Extraction (yellow): Identifies companies, products, people, places, events
Link Processing (orange): Classifies URLs and domains by type
Enrichment (purple): Calculates metrics, trends, and generates AI insights
Report Generation (blue): Creates interactive HTML reports with visualizations

🤝 Support

Create an Issue if you have any questions or problems

Issues

📄 License

Elastic License v2 (ELv2) - See LICENSE.md

In plain English:

✅ Run on your computer for any purpose - Personal use, commercial use, business use - completely free with no limits.
✅ Use, modify, and distribute freely
✅ View and audit all source code
✅ You control all costs - Only pay for AI API usage through OpenRouter/OpenAI/Perplexity as you consume it
❌ Cannot offer AICW as a hosted service to others

🔔 Schedule a Report

To create a report for the current date, run "project-build" pipeline and it will run full pipeline for the current date. The new report will be created and will indicate trend (growing, declining, stable) for every item where possible using newly captured data and data from previous reports.

If want to schedule capturing new data and generating a report with trends to be sent weekly or monthly to your inbox automatically, please visit https://aichatwatch.com/cloud for more information.

Contacts

Website
Email: aichatwatch@gmail.com