agentleFS
Sign inStart free
danishashkodanishashko/geo-aeo-tracker/cli/llms.txt

geo-aeo-tracker / cli

llms.txt328 starsChanged 28 days agoMITView on GitHub

CLI that measures how AI answer engines (ChatGPT, Perplexity, Gemini, Google AI Mode) mention, recommend and cite a brand, audits pages for AI search readiness (GEO/AEO), and returns Google rank. Runs with npx -y geo-aeo-tracker. Non-interactive; every command takes --json.

  • Reads credentials

The file

We split this llms.txt into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Start free: engines, then audit on the user's site.

    line 20
  2. Distinctive:

    Use buyer-style prompts ("What is the best CRM for a small business?"), not the brand name.

    line 21

≈ “Common” is an estimate: at least 1 in 5 of the 3,592 llms.txt files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
llms.txt
# geo-aeo-tracker

> CLI that measures how AI answer engines (ChatGPT, Perplexity, Gemini, Google AI Mode) mention, recommend and cite a brand, audits pages for AI search readiness (GEO/AEO), and returns Google rank. Runs with `npx -y geo-aeo-tracker`. Non-interactive; every command takes `--json`.

## Discover

- `npx -y geo-aeo-tracker help --json`: every command, flag, default, cost, duration, JSON output field and exit code as one JSON document. Read this before anything else.
- `npx -y geo-aeo-tracker help <command>`: the same for one command, as text.

## Commands

- `engines --json`: free. Engine ids (`chatgpt`, `perplexity`, `gemini`, `google_ai`) and known outages. Check before `check`.
- `audit <url> --json`: free, no key. 20 checks (llms.txt, AI crawler access, sitemap, JSON-LD, FAQ schema, Open Graph, answer-first content, headings, rendering) and a 0-100 score. Each failed check has `detail` saying how to fix it.
- `check "<buyer question>" --brand <name> [--competitor a,b] [--website domain] [--engine id] [--country CC] --json`: asks each engine the prompt in parallel and returns per-engine `visibilityScore` (0-100), `sentiment`, `brandMentions`, `competitorMentions`, `websiteCited`, `sources` and the full `answer`, plus `summary.averageVisibility`. Bills Bright Data once per engine; 30 s to 5 min.
- `serp "<keyword>" --target <domain> --json`: Google organic results and `targetRank`. Bills Bright Data once.

## Rules for agents

- Needs `BRIGHT_DATA_KEY` in the environment (or `.env` in the working folder) for `check` and `serp`. Never ask the user to paste it into chat.
- Start free: `engines`, then `audit` on the user's site. Run one `check` before batching many; each costs money.
- Use buyer-style prompts ("What is the best CRM for a small business?"), not the brand name.
- With `--json`, stdout is exactly one JSON document, also on errors: `{"error":{"type":"usage"|"failed","message":"...","exitCode":2}}`. Progress lines go to stderr.
- Exit codes: 0 ok, 1 failed (no engine answered, page unreachable), 2 bad usage or missing key (nothing billed), 3 below `--min-score`.
- An engine with `status: "error"` and a `note` is a known outage at Bright Data; report it and leave that engine out, do not retry in a loop.

## Docs

- [README](https://github.com/danishashko/geo-aeo-tracker/tree/main/cli#readme)
- [Dashboard and MCP server](https://github.com/danishashko/geo-aeo-tracker)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free