Skip to content
Cited·Index
Sign inSubmit a tool

auto-geo

Summary

auto-geo is an open-source, MIT-licensed command-line GEO engine that audits pages for citation readiness, generates and rewrites them, and measures whether AI search engines cite your domain. Tracked prompts and saved check runs are plain files committed in your own repository.

Not for
non-technical marketing teams that want a hosted dashboard, Google AI Overviews monitoring, or a done-for-you managed service
AI engines it tracks

The AI engines auto-geo tracks

  • ChatGPT: on every plan
  • Gemini: on every plan
  • Perplexity: on every plan
  • AI Overviews: not tracked
  • Claude: on every plan
  • AI Mode: not tracked
  • Copilot: not tracked
  • Grok: on every plan
  • DeepSeek: not tracked
  • Meta AI: not tracked

5 on every plan.

From
Free
Open source
YesMIT
API
Yes
Category
GEO tools
Best for
Developer-led teams or solo marketers
Screenshot of auto-geoCaptured · citedindex.com
Visit auto-geo CompareFind alternatives
Yours? Claim it free

Is auto-geo yours?

Owners claim at no cost, then see this page's readers and correct its details.

Claim this listing

What it does

Monitors AI answers
YesThe check command asks real grounded-search engines per tracked prompt and records whether your domain appears among the citations.For each prompt, ask a real AI search engine and report whether your domain is among the citations.github.com/shadowresearch/auto-geo
Optimises your content
YesThe doctor command audits any live URL against eight citation-readiness checks and returns a ranked list of fixes for that page.Run it on any URL — yours, a competitor's, every page in your sitemap — and get a structured report on the citation signals AI engines look for.github.com/shadowresearch/auto-geo
Generates content
YesThe write command generates whole publish-ready pages from target queries, validated against its strict seven-block GEO schema.Give it your domain and the queries you want to be cited for; get back validated, publish-ready JSON files — one structured page per querygithub.com/shadowresearch/auto-geo
Publishes to your site
NoIt emits renderer-agnostic JSON and leaves delivery to you: the docs tell the operator to publish through their own pipeline, with no CMS connector documented.publish the JSON through the user's own pipelinegithub.com/shadowresearch/auto-geo/blob…
Technical AEO
YesIt audits and derives Article and FAQPage JSON-LD from page structure and ships an llms.txt index of its own; no crawler-directive handling is documented.Was Partly before our last check.Schema-derived. Article + FAQPage JSON-LD emitted from structure, not hand-written.github.com/shadowresearch/auto-geo
Done-for-you service
NoNo people-delivered service: this is software you install yourself from npm, Homebrew or MacPorts and run on your own machine.Was Partly before our last check.Gemini CLI is an open-source AI agent that brings the power of Gemini directly into your terminal.github.com/google-gemini/gemini-cli

Answer engines

ChatGPT
YesAn OpenAI adapter queries the Responses API with the web search tool, and the geo-audit output format labels that provider chatgpt.Uses the Responses API with the web_search_preview tool.github.com/shadowresearch/auto-geo/blob…
Perplexity
YesPerplexity is the default engine, called through Sonar chat completions, with citations read from the response's search results field.Uses Perplexity's Chat Completions API with sonar (or sonar-pro).github.com/shadowresearch/auto-geo/blob…
Gemini
YesGoogle is named as one of the foundation model providers behind the CLI, and Gemini models appear in the per-plan model table.Copilot CLI supports models from multiple foundation model providers, such as Anthropic, Google, and OpenAI.github.com/features/copilot/cli
Google AI Overviews
NoThe documented engine list for the check command names five adapters only; Google AI Overviews is not one of them, so no tier queries it.Was Partly before our last check.Engines: perplexity (default), openai, anthropic, gemini, xai (alias grok), or --engine all — which runs every engine whose API key is set and…github.com/shadowresearch/auto-geo/blob…
Google AI Mode
NoThe documented engine list for the check command names five adapters only; Google AI Mode is not one of them, so no tier queries it.Was Partly before our last check.github.com/shadowresearch/auto-geo/blob…
Claude
YesClaude is the default model in the CLI and further Claude versions are selectable with the /model command.By default, copilot utilizes Claude Sonnet 4.5. Run the /model slash command to choose from other available modelsgithub.com/github/copilot-cli
Microsoft Copilot
NoThe documented engine list for the check command names five adapters only; Microsoft Copilot is not one of them, so no tier queries it.github.com/shadowresearch/auto-geo/blob…
Grok
YesGrok is covered through the xAI adapter, which turns on Live Search and reads the flat citations array the chat endpoint returns.Uses xAI's OpenAI-compatible chat completions endpoint with the search_parameters extensiongithub.com/shadowresearch/auto-geo/blob…
DeepSeek
NoThe documented engine list for the check command names five adapters only; DeepSeek is not one of them, so no tier queries it.github.com/shadowresearch/auto-geo/blob…
Meta AI
NoThe documented engine list for the check command names five adapters only; Meta AI is not one of them, so no tier queries it.github.com/shadowresearch/auto-geo/blob…
Coverage in the vendor’s words
YesFive engines ship as adapters with no plan gating at all; each needs your own API key, and engines whose key is unset are skipped rather than failed.github.com/shadowresearch/auto-geo
Reads answers as a user sees them
NoEvery engine is reached through its provider's grounded-search API with your own key, not by reading the consumer apps a person would use.check calls real grounded-search APIs. Provision the key for the engine you want to use as an environment variablegithub.com/shadowresearch/auto-geo/blob…

What it reports

Names the cited sources
YesEach result lists the exact cited URLs with their rank and the total citation count, plus the raw source list even for uncited queries.For each cited URL, rank is its 1-based position in the engine's citation list, and totalCitationsForQuery is the denominator.github.com/shadowresearch/auto-geo/blob…
Competitor tracking
PartlyYou can audit a competitor's page structure with doctor, but a check run measures one required domain at a time, so rival brands are not watched side by side in one view.--domain <d> | Bare domain (shadow.inc) or full origin (https://shadow.inc). Required.github.com/shadowresearch/auto-geo/blob…
Brand sentiment
NoNo sentiment or attribute analysis: the run reports coverage, cited URLs and rank, and can print the engine's raw answer text, leaving any reading of tone or attributes to you.Render the engine's natural-language answer under each query in human output. preview (default, ~3 sentences, ~400 chars), full, or none.github.com/shadowresearch/auto-geo/blob…
Recommendations
YesDoctor returns the failing checks ranked by citation impact with concrete fix text, and fix regenerates the page so it passes all eight checks.The fix list is ranked by citationImpactRank — the SOP-derived ordering of which signals carry the largest measured citation lift.github.com/shadowresearch/auto-geo/blob…
AI crawler analytics
NoNothing ingests server logs or reports AI bot hits; the tool keeps no server, account or database, and its own fetches are the only crawling it does.Everything is file-based and committable — tracked prompts, check history, config. No server, no account, no database. One install away:github.com/shadowresearch/auto-geo
Suggests prompts to track
YesA discover subcommand has an LLM propose high-intent queries from your homepage and existing set; no search-volume or ask-frequency figures are published.discover proposes them: it fetches your homepage for grounding, shows the model what you already track, and asks for N new high-intent queriesgithub.com/shadowresearch/auto-geo/blob…
Wrong answers about you
NoThe run records only whether your domain was cited, which URLs and at which rank; it never evaluates whether the engine's claims about your brand are accurate.For each prompt, ask a real AI search engine and report whether your domain is among the citations. This is the ground truth doctor predicts.github.com/shadowresearch/auto-geo

Plans and signing up

A month at 100 prompts
Yes$0 a month on Open source (MIT) for 100 tracked prompts. Prompts as the vendor counts them. Which AI engines check each one, and how often, depends on the plan.Audit, generate, fix, and track the pages large language models cite — one CLI, file-based, MIT.github.com/shadowresearch/auto-geo
A month at 500 prompts
Yes$0 a month on Open source (MIT) for 500 tracked prompts. Prompts as the vendor counts them. Which AI engines check each one, and how often, depends on the plan.github.com/shadowresearch/auto-geo
Prompt or credit allowance
YesThere is no plan allowance at all: tracked prompts are lines in a committed text file, and your only ceiling is the provider API spend each query costs..auto-geo/prompts.txt — plain text, one prompt per line, # comments allowed.github.com/shadowresearch/auto-geo/blob…
Projects or brands
YesConfig is resolved by walking up from the working directory, so separate repositories or packages each carry their own domain, prompt set and check history.Was Partly before our last check.auto-geo.config.json (walks up from cwd — monorepo-friendly)github.com/shadowresearch/auto-geo
Countries
NoNo country, locale or market flag exists in the documented command surface, and the page audit is explicitly English-only, so queries run wherever each provider defaults.English-only. The capitalization-based entity heuristic and the hedge-phrase regex assume English prose.github.com/shadowresearch/auto-geo/blob…
How often it checks
YesRuns are on demand — a manual command, a CI gate or a scheduled job you wire up yourself — rather than a vendor-run daily or weekly cadence.Was Partly before our last check.Use it as a CI gate, a weekly scorecard, or a one-shot competitive audit.github.com/shadowresearch/auto-geo/blob…
Card required to start
NoNothing is billed, so no payment detail is ever requested: the vendor states plainly that no API key and no account are needed to run the server.No upload, no API key, no account.github.com/founden-bot/yap-mcp
Self-serve signup
YesInstallation is a single npx command or a downloaded bundle, with no sales contact, sign-up or key issuance anywhere in the documented path.No. Yap MCP needs Node.js 20 or newer and nothing else. It is MIT licensed; the speech model is CC-BY-4.0.yapapp.cc/mcp

Access and integrations

API
YesProgrammatic mode drives the agent from your own scripts with a single prompt argument, and it can authenticate with a fine-grained personal access token.Was Partly before our last check.To use the CLI programmatically, include the -p or --prompt command-line option in your command.docs.github.com/en/copilot/concepts/cop…
MCP server
YesIt ships with GitHub's MCP server enabled and connects to custom MCP servers, so it acts as an MCP client rather than a server other agents call.Was No before our last check.the coding agent ships with GitHub's MCP server by default and supports custom MCP servers to extend capabilitiesgithub.com/github/copilot-cli

For a larger team

Single sign-on
PartlyThe plan comparison marks SAML SSO authentication as not included on Free, Pro, Pro+ and Max, so it is a team-plan concern rather than an individual one.SAML SSO authentication. Free plan: Not included. Pro plan: Not included. Pro+ plan: Not included. Max plan: Not includedgithub.com/features/copilot/plans
Security certifications
YesGitHub states Copilot supports GDPR compliance through a data protection agreement; SOC 2 or ISO 27001 certificates are not named on the pages read.Yes. GitHub and customers can enter a Data Protection Agreement that supports compliance with the GDPR and similar legislation.github.com/features/copilot/plans

Checked · Named in 3 of 242 AI answers, October check · 8 facts changed · How we check · Editorial policy

Changed since our last check

  • Technical AEO: now Yes. It was Partly when we checked on .
  • Done-for-you service: now No. It was Partly when we checked on .
  • Google AI Overviews: now No. It was Partly when we checked on .
  • Google AI Mode: now No. It was Partly when we checked on .
  • Projects or brands: now Yes. It was Partly when we checked on .
  • How often it checks: now Yes. It was Partly when we checked on .
  • API: now Yes. It was Partly when we checked on .
  • MCP server: now Yes. It was No when we checked on .

Plans and pricing

Price and terms

Free plan
YesYesThe whole tool is MIT-licensed and installed from npm, Homebrew or a standalone binary with no account or card
The terms in full

you supply and pay for your own engine API keys (Perplexity, OpenAI, Anthropic, Gemini, xAI), and the docs print per-run cost estimates such as roughly $0.012 for a three-query check.

Card to start
YesNot required
Sign up without sales
YesYes

Checked .

Works with

  • ChatGPT
  • Perplexity
  • Gemini
  • Claude
  • Grok

For AI agents

MCP server
Yes
API
Yes
AdYour company across CitedIndexAdvertise here

Alternatives

  • geoSurge

    Tracks how ChatGPT, Claude and Gemini describe your brand, free on your own AI subscriptions.

  • Rankscale

    Credit-based AI visibility platform tracking brand mentions, citations and sentiment across 17+ AI engines.

Screenshots

  • Pricing page of auto-geo
    Pricing page, github.com, captured
  • Docs of auto-geo
    Docs, github.com, captured

Filed under

GEO toolsAI visibility monitoringChatGPTClaudeContent optimizationGeminiPerplexity

Sources and updates

Last changed

Sources20

Repository30 August 2026

Licence
MIT
Stars
23
Forks
0
Last push
1 August 2026

Updates4

  • Features and coverage · Free tier · Plan limits · Terms
  • Features and coverage · Free tier · Pricing
  • Features and coverage · Free tier
  • Features and coverage · Free tier

Is this yours? AI assistants like ChatGPT and Claude opened CitedIndex pages 2,286 times in September. Claim it free and edit it in your dashboard, or feature it in GEO tools.

The digestFree

Who's winning AI citations.

The AI visibility tools we add and the guides we write, by email, at most once a week.

Last issue · No. 1 · 13 October 2026GEO and Content Tools for Agencies, plus 3 new tools

142 tools trackedOne-click unsubscribe