Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.cognerd.in/llms.txt

Use this file to discover all available pages before exploring further.

GEO Assets (Generative Engine Optimization Assets) are a set of machine-readable files that give AI crawlers, agents, and language models a structured, authoritative description of your brand. When AI systems can read these files directly, they produce more accurate, more frequent, and more positively framed mentions of your brand in AI-generated answers. CogNerd generates all seven files in a single async job by crawling your website, enriching the content it finds, and packaging everything into a ready-to-deploy ZIP archive.
Generating GEO Assets costs 20 credits per run. The job runs asynchronously — you receive a job ID immediately and can poll for completion or wait for the in-dashboard notification.

The seven GEO files

CogNerd produces exactly seven files in each GEO Assets job:

llms.txt

A plain-text, machine-readable brand summary written specifically for AI agents. Describes who you are, what you do, your key products, and how AI models should represent your brand.

robots.txt

An AI-aware crawl rules file. Controls which AI crawlers can access which parts of your site, and sets directives that comply with the crawling policies of major AI platforms.

sitemap.xml

A page index formatted for AI crawlers as well as traditional search engines. Prioritizes your highest-value pages and signals content freshness dates.

site.jsonld

A schema.org JSON-LD file that defines your site as a structured entity — including its name, URL, description, and relationship to your organization.

organization.jsonld

A schema.org JSON-LD file that describes your brand as a legal and commercial entity — including your logo, founding date, industry, social profiles, and contact information.

website.jsonld

A schema.org JSON-LD file that describes your website and its search action — enabling AI agents to understand how to navigate and query your site programmatically.

entities.json

An entity graph that maps the key concepts, products, people, and topics associated with your brand. AI models use entity graphs to understand context and relationships.

Generate GEO Assets

1

Open GEO Assets

Navigate to Dashboard → GEO Assets.
2

Review your brand profile

GEO Assets generation uses your brand name, website URL, and scraped site content. Confirm these are correct in Settings → Brand Profile before generating. Inaccurate brand details produce inaccurate GEO files.
3

Click Generate

Click the Generate GEO Assets button. CogNerd starts the four-stage pipeline immediately and returns a job ID. You can stay on the page to watch progress, or navigate away — the job runs in the background.
4

Wait for completion

The four pipeline stages typically complete in 3–8 minutes depending on your website’s size. The dashboard shows live progress across each stage.
5

Download your files

When the job completes, click Download ZIP to get all seven files in a single archive. Deploy them to your web server as described below.

How the pipeline works

GEO Assets generation runs four stages in sequence:
1

Page discovery

CogNerd crawls your website’s sitemap and navigation to identify the most important pages — your homepage, product pages, about pages, and key blog content.
2

Content crawl

CogNerd fetches and extracts the text, metadata, and structured data from each discovered page. This content forms the factual basis for all seven generated files.
3

Enhanced asset generation

CogNerd uses the crawled content to generate each file — writing your llms.txt brand summary, constructing your JSON-LD schemas, building your entity graph, and producing the AI-aware robots.txt and sitemap.xml.
4

Final report

CogNerd assembles the final report and packages all seven files into a downloadable ZIP archive. The report includes a validation summary showing any schema errors or missing fields.

Deploy your GEO files

After downloading the ZIP, deploy each file to the correct location on your web server:
FileDeploy location
llms.txthttps://yourdomain.com/llms.txt
robots.txthttps://yourdomain.com/robots.txt
sitemap.xmlhttps://yourdomain.com/sitemap.xml
site.jsonldhttps://yourdomain.com/site.jsonld
organization.jsonldhttps://yourdomain.com/organization.jsonld
website.jsonldhttps://yourdomain.com/website.jsonld
entities.jsonhttps://yourdomain.com/entities.json
The three JSON-LD files can also be embedded as <script type="application/ld+json"> blocks in your page <head> if you prefer inline schema over separate file hosting.
After deploying, run a new Brand Monitor analysis 7–14 days later to measure the impact on your AI Visibility Score. GEO files typically show their strongest effect on citation rate and platform coverage.
If you already have a robots.txt or sitemap.xml on your site, review the generated files carefully before replacing them. The generated versions are optimized for AI crawlers but you should ensure they don’t accidentally block legitimate crawlers you rely on.

Understand each file

llms.txt is a plain-text file placed at the root of your domain, intended to be read by AI agents before they interact with your site. It describes your brand, your products, your intended audience, and how you want your brand represented in AI-generated content. Think of it as a machine-readable brand brief — concise, factual, and authoritative.
The standard robots.txt was designed for traditional search engine crawlers. CogNerd’s AI-aware version adds directives for AI crawlers (such as GPTBot, ClaudeBot, Google-Extended, and others) that control which parts of your site they can index and at what rate. This lets you grant AI crawlers access to high-value pages while protecting content you don’t want included in AI training data.
These three JSON-LD files follow the schema.org vocabulary. organization.jsonld is the most important — it defines your brand as a named entity with verifiable properties (logo URL, founding year, industry, social media profiles). This helps AI models confidently associate facts about your brand with the correct entity rather than confusing you with similarly named brands.
The entity graph maps relationships between your brand and associated concepts — your product categories, leadership names, partner brands, geographic markets, and topics you publish about. AI models use entity graphs to answer nuanced questions like “who is the CEO of [your brand]?” or “what industries does [your brand] serve?” accurately.
Regenerate whenever you make significant changes to your brand — rebranding, launching new products, expanding to new markets, or adding or removing leadership. As a baseline, regenerating quarterly keeps your GEO files aligned with your current brand state. Each regeneration costs 20 credits.

Check job status

If you navigate away during generation, you can check your job status at Dashboard → GEO Assets → Job History. Each job entry shows the current stage, elapsed time, and final status (pending / processing / complete / failed). If a job fails, the dashboard shows which stage encountered an error. Common causes are websites that block crawlers or have no accessible sitemap. Update your robots.txt to allow CogNerd’s crawler, then re-run the job.