How to Check If AI Systems Can Find and Cite Your Site (in 5 Minutes)
A new CLI tool called LLMScout runs 21 technical SEO and GEO checks to evaluate your website readiness for AI systems.
Stock photo for illustration only, not from the actual event
- LLMScout executes 21 comprehensive checks covering Technical SEO and GEO.
- Position-1 click-through rate on AI Overview keywords dropped from 7.3% to 1.6%.
- Separately reports crawler directives for model training bots versus live retrieval bots.
- Available via npm and pip distributions with minimal or zero runtime dependencies.
As artificial intelligence systems increasingly dominate information discovery, position-one click-through rates on keywords triggering an AI Overview plummeted from 7.3% in December 2023 to 1.6% in December 2025, according to data from Ahrefs. If an AI engine fails to discover or cite your website, that lost traffic will not recover on its own, making immediate site auditing essential.
A Dev.to article co-authored by Rudrendu Paul and Sourav Nandy introduces LLMScout, a cross-platform, zero-dependency Command Line Interface designed to run 21 distinct checks against any target website. The suite includes 12 checks dedicated to technical SEO fundamentals and 9 focused strictly on Generative Engine Optimization (GEO) signals tailored for AI platforms.
Stock photo for illustration only, not from the actual event
Generative Engine Optimization (GEO) is emerging as a critical successor to traditional SEO practices as user behavior shifts from browsing standard search result links to consuming AI-synthesized summaries. CLI-based auditing tools allow engineering teams to seamlessly integrate compliance and discoverability checks directly into CI/CD pipelines before code reaches production environments.
A standout feature of the utility is its ai-crawler-directives check, which independently evaluates all seven major AI bots. Blocking GPTBot, which is utilized for model training, does nothing to prevent OAI-SearchBot from fetching and citing pages for live ChatGPT retrieval. Because these bots are independently configurable via robots.txt, LLMScout maintains this distinction rather than collapsing them into a single warning.
Installation is supported through two primary package managers:
- npm install -g llmscout-cli: Requires Node 18+ and pulls in precisely two runtime dependencies (cheerio and commander) without requiring Python or headless browsers.
- pip install llmscout-cli: A pure standard library Python version featuring zero runtime dependencies while executing the exact same 21 checks.
Once installed, users can initialize a configuration file and execute audits using standard commands:
llmscout init ./my-site --site-url https://yourdomain.com
llmscout check ./my-site
The audit yields PASS, WARN, or FAIL verdicts paired with built-in remediation advice directly within the output lines. Developers can also pass the --json flag to integrate the tool into automated CI pipelines, causing builds to fail automatically upon encountering critical errors.
Source: Dev.to
Found something wrong in this article? Report an issue with this article
Comments
Leave a Comment