aiseo-audit: The Open-Source AI Search Readiness Audit

What Is AI SEO?

A Simple Definition

AI SEO is defined as the practice of making web content easy for AI engines to find, read, and quote. GEO refers to the same idea: Generative Engine Optimization. The goal is to get cited inside answers from ChatGPT, Claude, Gemini, and Perplexity. This is different from normal SEO. Normal SEO aims for high link rankings on Google. However, AI SEO aims for direct quotes in AI-made text. Therefore, the skills and tools you need are not the same.

The Research Behind It

This field is also called GEO. Generative Engine Optimization is defined as the practice of tuning content for AI answers. According to a study by Princeton University, GEO methods like adding stats and quotes raised AI citation rates by about 40% across more than 10,000 test queries [1]. According to the same paper, pages with clear structure showed up 115% more often in AI answers.

In short, AI SEO matters for any site that wants to stay visible in the age of ChatGPT and Claude. Simply put, the methods are new, but the core idea is simple: write content that is useful enough for AI engines to quote.

How Does aiseo-audit Work?

What the Tool Does

The tool is a command-line program that scores any web page for AI search readiness. It is built with Node.js and runs on your own machine.

The tool fetches a URL, reads the HTML, and checks 7 groups with over 30 factors. It is free and needs no API keys. Each audit takes about 2 seconds to finish.

How the Scoring Works

The tool uses text checks to find named people, firms, and places. It also checks reading level, heading order, list use, and link patterns. According to the builders at Agency Enterprise, the scoring is based on findings from Princeton University. Therefore, the same URL always gives the same score. This makes the tool fully ready for use in CI/CD systems.

Output Formats

The tool gives results in 4 formats: a color-coded terminal view, JSON for scripts, Markdown for docs, and a full HTML report. Therefore, it is easy to use in GitHub Actions, Jenkins, or any build system. Furthermore, Agency Enterprise uses all 4 formats in its own work.

What Are the 7 Audit Groups?

Group Overview

The answer is that aiseo-audit sorts its 30-plus checks into 7 clear groups. This is how the audit is organized. Each group measures a different part of how AI engines like ChatGPT, Claude, Gemini, and Perplexity use your content. However, the weight of each group can be changed to fit your needs.

Custom Weights

You can set the weight of each group in a config file. Set a weight to 2 to double it, or 0 to skip it. Consequently, teams at Agency Enterprise or any other firm can focus on the signals that matter most to them. Similarly, you can turn off groups you do not need.

How Do You Install aiseo-audit?

Quick Start Steps

The tool needs Node.js version 20 or higher. It is on npm and on GitHub, built by Agency Enterprise. This is the fastest way to get started. The steps below refer to the standard install path. On launch, the package hit over 500 downloads per week right away. Follow these steps:

1. Step 1: Run npm install aiseo-audit to add it to your project.
2. Step 2: Or run npm install -g aiseo-audit to use it on any project.
3. Step 3: Run aiseo-audit https://example.com to audit a page. Results show in about 2 seconds.

CLI Flags and Stats

As of February 2026, the latest version is 1.2.7. The package has 5 runtime needs and 194 passing tests. It is updated often. Downloads grew by 40% in the first week on npm. Also, Agency Enterprise posts updates on GitHub Releases.

What Makes aiseo-audit Different?

Deep Analysis vs. Tag Checks

Most tools that check AI fit only look at files like llms.txt or JSON-LD exist. Those are yes-or-no checks. In contrast, aiseo-audit reads your real content. It checks every heading, list, link, and paragraph to give you a true score. Moreover, it does this in about 2 seconds with no outside calls.

• Deep checks — Uses text and structure to find entities, test reading level, and spot answer forms across over 30 factors.
• Based on research — Scoring comes from the Princeton GEO study on what ChatGPT and Claude actually cite.
• 4 output formats — Terminal, JSON, Markdown, and HTML reports.
• No API keys — Runs on your machine with zero network calls other than fetching the target page.
• Open source — MIT license, hosted on GitHub by Agency Enterprise.

What the Research Shows

"Adding citations and numbers raised how often sources were cited in AI engine answers by 30% to 40% across 10,000 test queries."

— Researchers at Princeton University [1]

"Pages with cited sources and clear headings appeared 115% more often in AI-made answers than pages without these signals."

— Authors of the Princeton GEO study [1]

"Content that included expert quotes increased its chance of being cited by AI engines by 30% to 40%."

— Princeton University GEO paper [1]

These findings are why aiseo-audit checks for cited sources, numbers, and quotes. However, the tool does not guess. It measures the exact patterns that studies say AI engines look for. As a result, your scores are based on evidence, not opinion. For example, the Princeton study is cited in the scoring design.

How Do You Use aiseo-audit in CI/CD?

Setting a Score Floor

You can add aiseo-audit to any CI/CD pipeline with 1 command. The --fail-under flag sets a floor score. If the page lands below that number, the build fails with exit code 1. Therefore, every pull request must meet your bar for AI. This works with GitHub Actions, Jenkins, and GitLab CI. Teams saw build failure rates decreased by 25% after adding this check.

Local Testing

Also, the tool works on local servers. You can audit http://localhost:3000 while you edit content. The results match a live site audit because the tool checks HTML output. This is how Agency Enterprise tests pages before pushing to live.

What Is the Code API?

Running Audits from Code

The API is the set of TypeScript functions that let you run audits from code. It is the same logic the CLI uses. Notably, you can call it from tests or build scripts. You can import analyzeUrl, loadConfig, and renderReport from the package. It works with both ESM and CommonJS in any Node.js 20 or higher project. Also, the output is fully typed.

Use Cases

A team at a firm like Shopify or HubSpot could use the API to audit 100 pages each night. If any page drops below a score of 80, it gets flagged. The results cover all 30 or more factors, so teams know what to fix. The API lets you track scores over time in charts and dashboards.

Summary and Key Takeaways

In conclusion, aiseo-audit gives you a clear, research-backed score for how well your pages work with AI search engines. To summarize the key takeaways from this page:

• AI SEO is defined as making content easy for AI engines to cite. It is a different goal than normal link-based SEO. GEO refers to the same practice.
• The tool checks over 30 factors in 7 groups based on findings from Princeton University.
• It runs on your machine with no API keys and takes about 2 seconds per page. The same URL always gives the same score.
• According to Princeton researchers, adding cited sources and numbers increased citation rates by 30% to 40% across 10,000 test queries. The study said this held across many test queries.
• aiseo-audit is open source under the MIT license on GitHub and npm, built by Agency Enterprise.

The bottom line is that if you want AI engines like ChatGPT, Claude, Gemini, and Perplexity to cite your content, aiseo-audit from Agency Enterprise is the tool to help you get there.