Track weekly whether Gemini recommends your product with one Python script on GitHub Actions and the free API tier
A developer built a one-file Python script that asks Gemini five buying-intent questions every Monday, checks whether your brand and up to ten competitors are mentioned and in what position, and saves an HTML report as a GitHub artifact.
Evidence: The source shows its work: steps, screenshots, code or data you can inspect.
The business problem
Founders want to know whether AI assistants recommend them for buyer questions, but paid AI-visibility trackers start around $29 a month.
What was tried
1. Fill in a config file with your brand, up to 10 competitors and 5 buying-intent prompts (for example, best alternatives to a known competitor). 2. Store a free-tier Gemini API key as a GitHub secret. 3. A GitHub Actions workflow runs the standard-library Python script every Monday at 08:00 UTC. 4. The script sends each prompt to Gemini, records whether each brand is mentioned, its position when the answer is a list, and which competitors appear alongside it. 5. It writes an HTML report with a snippet and the full answer; reports are kept as workflow artifacts so you can compare week over week.
What was reported (not reported)
Only a demo run is shown: using Linear as the brand with Jira, Asana and ClickUp as competitors, Linear was mentioned in all three test prompts, ranked first in two and second in one. The author reports running costs of $0 on Gemini's free tier. No results from the author's own product or from buyers are reported.
Limitations
Vendor: the author sells the source code for $19 one-time (rising to $29), with a public landing page and repo. Gemini only, so it says nothing about ChatGPT, Claude or Perplexity. Position parsing works only when the answer is a list; prose answers give a mention count only, and a brand nested under a competitor still counts as mentioned. One demo, published the same day, so no track record. Needs comfort with GitHub secrets and JSON. Free-tier limits may change.
What you need
A GitHub account, a Gemini API key, and a list of buyer questions and competitors.
Keep the prompt set fixed from week to week so changes reflect the answers, not your wording.
Sources
- DEV Community (98dkcc488b-glitch) ↗ Vendor case study, published October 25, 2026, main source
- GitHub (98dkcc488b-glitch/askvisible) ↗ Code repository, publication date unknown
Source published: October 25, 2026. Last reviewed here: October 25, 2026. Spot a mistake? Tell us.
Tools in this workflow
- GitHub Actions: Runs the script weekly and stores reportsTry GitHub Actions
- Google Gemini API: Answers the buyer questionsTry Google Gemini API
Get new workflows like this by email
A short selection of the most useful new workflows, including the ones that did not work. Free, and you can unsubscribe at any time.
Related workflows
Adding GPT-5 coaching, gamification and four onboarding rebuilds did not lift day-one retention in a solo habit app
A solo developer shared three years of habit-app data: an AI coaching chat converted well when found but only 2.4% of users found it, and day-one retention stayed at 4.2% through AI, gamification and onboarding changes.
Measure whether ChatGPT and Gemini recommend your SaaS: 25 unbranded buyer questions, three numbers and a fixed weekly set
A founder asked ChatGPT and Gemini 25 questions buyers ask before they know any brand, logged every brand and cited URL, and scored visibility share, citation share and average position. Their product was named in 1 of 50 answers; a competitor in 19.
Map which kinds of pages ChatGPT, Perplexity, Gemini and Google AI Overviews cite for your buyers' questions
A PR agency founder asked four AI assistants 15 buyer questions in three B2B categories, logged all 523 cited sources and sorted them by type, finding each assistant and question type leans on different pages.
Get a weekly Claude Code report that links shipped pull requests to KPI changes and scores the likely impact
A product lead connects Claude Code to Metabase and GitHub so each week it compares 5 to 10 KPIs with last week, maps the week's pull requests to them with a confidence score, and recommends next steps.