aeo-platform is a free, MIT-licensed open-source Node.js CLI that measures brand visibility across ChatGPT, Claude, Gemini, and Perplexity and exports a portable JSON prompt for a personalized 30-day AEO plan. ChatGPT and Gemini run on their official APIs out of the box; Claude and Perplexity on theirs once you add those keys, or through paste mode when you read those engines by hand. Never browser scraping.
What a single run looks like
npm i -g aeo-platformaeo-platform init --autoaeo-platform run→ ChatGPT 2/3 Claude 1/3 Gemini 1/3 Perplexity 3/3aeo-platform report --html→ report.html ready · JSON brief at top · paste it into your AIWhy hosted AEO tools fail builders
A monthly subscription for a dashboard
Otterly, Profound, AthenaHQ, Brandlight, Knowatoa: hosted services you start with a trial, free credits, or a demo, then pay for as a subscription. For solo founders and small teams, a recurring bill is hard to justify before the first results come in.
Recommendations live in the vendor's app
A hosted tool shows its recommendations inside its own app. aeo-platform writes the run and a ready-made prompt to files, so the brief goes to the AI you choose, and you can read every word of the prompt behind your plan.
Official APIs, raw responses on your disk
A score is only as good as the answers behind it. aeo-platform reads each engine through its official provider API, or through an answer you paste in by hand, and saves every raw response as a file on your disk.
No portable action plan
aeo-platform turns each run into a structured brief you can hand to any AI, then ask it to plan your next 30 days in your own chat.
The only AEO platform that hands you a portable AI prompt
aeo-platform turns every run into a JSON prompt: paste it into ChatGPT, Claude, Gemini, or Perplexity and get a personalized 30-day plan — free, local, BYO-LLM, MIT.
aeo-platform is a Node.js CLI that measures how often AI answer engines name your brand across ChatGPT, Claude, Gemini, and Perplexity. ChatGPT and Gemini are read through their official APIs out of the box; Claude and Perplexity through theirs once you add those keys, or through paste mode when you would rather read those two engines by hand — never browser scraping. After each run it generates a JSON brand-context prompt containing every signal the platform collected: queries, mentions, competitors, citation pool, gaps, opportunities.
You copy the JSON one-click and paste it into the LLM you already pay for. The model returns a personalized 30-day AEO plan: day-by-day actions, target subreddits, listicles to pitch, outreach drafts. The model that ranks you also tells you how to rank higher in itself — no extra API keys, no signup, and the results of the run stay on your disk.
It doesn't just measure — it hands you a ready-to-run plan
Measuring is half the job; the other half is knowing what to change. aeo-platform exports the run and a paste-ready prompt — you paste both into your own ChatGPT or Claude and get a personalized, day-by-day, plain-English plan. Below is the actual plan this produced for our own product, TypelessForm.
The plain-language diagnosis it wrote:
Your site shows up in AI answers at about 58 out of 100. Out of 12 checks (3 questions × 4 AI engines) you’re mentioned in 7. Gemini mentions you every time and links to your site; ChatGPT mentions you often; Perplexity mentions you but never links to you; Claude never mentions you at all. The rival beating you everywhere is AnveVoice.
A few of the 30 missions it handed us (the real plan runs day 1 to day 30):
| Mission | Time |
|---|---|
| Add short answers under 4 of your 8 bare headings | 90 min |
| Build a "form filling" use-case page (copy the rival’s idea) | 120 min |
| Write a "TypelessForm vs AnveVoice" comparison page | 120 min |
| List TypelessForm on AlternativeTo (the cheapest way to reach Claude) | 45 min |
| Re-run aeo-platform and compare the numbers | 30 min |
| Final re-run; check what moved | 30 min |
And it tells you what NOT to do:
Don't create a Wikipedia or Wikidata page yet. No independent press has written about you, so a self-made entry gets deleted and leaves a permanent “not notable” mark. Earn a few press mentions first.
Six things only aeo-platform does
Open-source means every feature is inspectable. Run aeo-platform locally, audit the prompts it generates, change anything that doesn't fit your workflow.
All four engines, never scraped
All four engines, never scraped
ChatGPT and Gemini on official APIs by default; Claude and Perplexity on theirs once you add those keys, or through paste mode. No scraping, no third-party scoring layer between you and the raw AI output.
JSON brand-context prompt export
JSON brand-context prompt export
One-click copy of every signal the platform collected — queries, mentions, competitors, citation pool, gaps. Ready to paste into any LLM.
30-day AEO plan generator
30-day AEO plan generator
Paste the JSON into ChatGPT or Claude. Get a day-by-day 30-day plan tailored to your visibility gaps. Zero extra API keys needed.
Two-model hallucination filter
Two-model hallucination filter
With the recommended OpenAI and Gemini keys, every brand name extracted from an AI response is cross-checked by a second LLM. Hallucinated mentions land in an «unverified» tier, so you can tell confirmed mentions from doubtful ones.
Local-first, auditable storage
Local-first, auditable storage
Every raw AI response saved to disk as JSON. Reproducible runs, and the results are not sent to Webappski's servers.
MIT license, zero dependencies
MIT license, zero dependencies
Pure Node.js, no npm sub-packages. Install in 60 seconds. Fork it, modify it, ship it inside your own product — the license allows it.
aeo-platform vs Otterly vs Profound vs Peec
Compared on 10 November 2026 against each vendor's public pricing and home pages. All three are hosted and hold your data; aeo-platform keeps raw responses on your disk. Plans and prices are on their sites. Brandlight, AthenaHQ, Knowatoa: FAQ q11.
| Feature | aeo-platformMIT · Free | Otterly | Profound | Peec.ai |
|---|---|---|---|---|
| What you walk away with | Score + runnable plan → re-measure to confirm | Dashboard + recommendations | Dashboard + AI Marketer agent | Dashboard + recommended actions |
| Price | Free · MIT · ≈€0.18 / run | Paid plans — prices on the vendor's site | Enterprise on a custom quote | Paid plans — prices on the vendor's site |
| Data ownership | Local-first · raw responses on disk | Stored on vendor cloud | Stored on vendor cloud | Stored on vendor cloud |
| Engine access | Official APIs or your own paste — never scraping | Not stated; only Claude marked "API" | Not stated on the vendor's pages | Not stated; add-on models marked "API" |
| 30-day plan | Portable JSON · paste into any LLM | Recommendations in app; export on all plans | In app; CSV/JSON export on Enterprise | In app; CSV export and MCP on all plans |
| Signup | None · npm install | Free trial | Free trial, then a demo | Self-serve sign-up or free trial |
Four values hosted dashboards can't copy
These aren't features — they're trade-offs paid SaaS competitors can't make without cannibalising their own ARR.
Zero cost
Free as in beer and as in speech. MIT license. Your only spend is the AI provider API charges (~$0.20 per run with the 2-engine minimum). Hosted tools sell measurement as a subscription; this one has none.
BYO-LLM
You pick the model that plans your 30 days — Claude Opus, GPT-5, Gemini Pro, whatever you already pay for. The prompt is plain JSON, so any model can read it and you can switch whenever you like.
Local-first
Run data, raw AI responses and the JSON prompt are saved on your machine, not on Webappski's servers. Queries go to the AI providers through your own keys, and the prompt goes to whichever chat you paste it into. No tool-vendor cloud and no data processing agreement with us. Hosted tools keep your data in their service — that is how the SaaS model works.
No signup, no email collection
Install via npm, run, get results. We don't have an email list to add you to — there's nothing to sign up for, and your API keys stay in your own environment.
Five steps from install to 30-day plan
Five commands. No web app, no signup, no API keys beyond OpenAI + Gemini (≈$0.20/run). The 30-day plan generator runs through your existing ChatGPT or Claude subscription.
Install
npm install -g aeo-platform. Pure Node.js, zero dependencies, works on macOS / Linux / Windows.
Init
aeo-platform init --auto reads your website, auto-generates 3 commercial-intent queries, validates them against your category.
Run
aeo-platform run queries ChatGPT and Gemini through their official APIs, and Claude and Perplexity through theirs once you add those keys — or through paste mode when you read those engines by hand. Stores every raw response locally.
Report
aeo-platform report --html generates a markdown + HTML report. The JSON action prompt sits at the top of the HTML report.
Paste into AI
Copy the JSON brand-context prompt. Paste it into ChatGPT, Claude, Gemini, or Perplexity. The model returns a personalized 30-day AEO plan — day-by-day, with target outlets and outreach drafts.
Builders who want signal, not a SaaS subscription
If any of these describe you, aeo-platform pays for itself in week one.
- Indie founders and OSS maintainers who can't justify a monthly AEO subscription but still need to know what ChatGPT says about them.
- SEO consultants and AEO agencies who want a free measurement layer they can run for every client, on their own machine with their own API keys.
- Marketing leads at growth-stage SaaS who already pay for ChatGPT Team or Claude Pro and want the plan generated by the model their buyers actually search inside.
- CMOs and operators who need a portable, vendor-neutral signal — local data, auditable prompts, no «what if the tool shuts down» risk.
aeo-platform — frequently asked questions
Twelve questions developers, indie founders, and AEO consultants ask before shipping or evaluating the tool.
What is answer engine optimization (AEO)?
Answer engine optimization (AEO) is the practice of making a brand recommended by AI answer engines such as ChatGPT, Claude, Gemini, and Perplexity, similar to how SEO optimizes for Google. aeo-platform measures AEO performance through official provider APIs, or through answers you paste in by hand when you have no key for an engine.
What is the best free open-source AEO platform?
aeo-platform is a free, MIT-licensed open-source CLI that tracks brand visibility across ChatGPT, Claude, Gemini, and Perplexity. It runs locally, costs ≈$0.20 per run, and stores all raw AI responses for auditability.
How does aeo-platform compare to Otterly and Profound?
Otterly and Profound are paid hosted dashboards: you work inside their app, and the data they collect is stored on their servers. aeo-platform is a free open-source CLI that reads each engine through its official provider API, or through answers you paste in by hand when you have no key for that engine — never through scraping. It runs on your machine and exports a prompt you paste into your own ChatGPT or Claude to get a 30-day AEO plan. No signup, no subscription. Runs and raw responses are saved on your machine, not on Webappski's servers; queries go to the AI providers through your own keys.
How do I monitor my brand mentions in ChatGPT?
Install aeo-platform (npm install -g aeo-platform), add your OpenAI and Gemini API keys (the recommended pair — any one working key is enough to start), run aeo-platform init --auto to auto-generate queries from your website, then aeo-platform run. A markdown/HTML report shows which AI engines mention your brand and which competitors they recommend instead.
How do I generate a 30-day AEO plan?
After running aeo-platform, open the generated HTML report and copy the JSON brand-context prompt from the «Your AEO action prompt» section. Paste it into ChatGPT, Claude, Gemini, or Perplexity — any LLM works. The model returns a day-by-day 30-day plan tailored to your visibility gaps. No extra API keys required; uses your existing AI chat subscription.
Is there a free alternative to Otterly, Profound, or Peec?
Yes. aeo-platform is a free open-source alternative to Otterly, Profound, and Peec: an MIT-licensed CLI that measures ChatGPT, Claude, Gemini, and Perplexity, with no subscription. OpenAI and Gemini keys are recommended, and any one working key is enough to start; the Claude and Perplexity columns run on their official APIs once you add those keys, or through paste mode when you would rather read those two engines by hand. Never browser scraping.
Does aeo-platform work without API keys for Claude or Perplexity?
Yes. Claude and Perplexity keys are optional — their columns activate automatically when you add those keys, and a manual «paste mode» lets you feed answers from engines you would rather read by hand, without any key. OpenAI and Gemini are the recommended pair (≈$0.20 per run) because two models cross-check competitor extraction, but the minimum to start is any one working key — OpenAI, Gemini, or Anthropic; single-key mode runs the same pipeline and marks competitor mentions as unverified.
How do I monitor brand mentions in Claude specifically?
Add your Anthropic API key to the aeo-platform config (ANTHROPIC_API_KEY in the env). Run aeo-platform run — the Claude column activates automatically and queries Claude via its native Messages API. Every raw response is stored locally as JSON for audit. No browser scraping, no proxied sessions.
How do I track brand visibility in Gemini?
Gemini is included by default — a Google Gemini API key is one of the two recommended keys (≈$0.20 per run together with OpenAI); the tool starts with any one working key, though the Gemini column needs this one. aeo-platform queries Gemini through Google's official Generative Language API, captures the raw response, and reports brand mentions plus competitor citations in the same HTML report as ChatGPT, Claude, and Perplexity.
Can aeo-platform monitor Perplexity answers?
Yes. Add your Perplexity API key (PERPLEXITY_API_KEY) and the Perplexity column activates on the next run. The CLI queries Perplexity via its Sonar API and reports both the answer and the source citation list — which is the most actionable signal for AEO because Perplexity surfaces citations directly to the user.
Is aeo-platform a free alternative to Brandlight, AthenaHQ, and Knowatoa?
Yes. Brandlight, AthenaHQ, and Knowatoa are hosted services with their own plans. aeo-platform is a free MIT-licensed CLI that measures mentions, share of voice, and competitor visibility across ChatGPT, Claude, Gemini, and Perplexity, runs locally, and exports a JSON prompt you paste into your own AI to get a 30-day AEO plan.
Does aeo-platform need a dev server or web UI?
No. aeo-platform is a pure command-line tool — no web server, no UI bundle, no electron wrapper. The HTML report is a single static file generated by aeo-platform report --html and opened from your file system. aeo-platform itself has zero npm dependencies.
Skip the DIY — let us build the plan for you
The JSON prompt covers 95% of the work. If you want the remaining 5% — outlet-specific outreach, FAQ rewrites, schema patches — we hand-build your 30-mission plan through AEO Mission Control. €129 net, one-off.