We open-sourced 4 repos for AI search and small-business automation

A few weeks ago we shipped free browser tools on Hugging Face. The most common follow-up was "can I use this in my own scripts?" So we put the data, policies and checkers on GitHub, plus two resources we keep sending to clients. Everything is free to use and adapt.

1. ai-crawler-robots-txt

https://github.com/313Automations/ai-crawler-robots-txt

"AI crawlers" are not one thing. Training crawlers (GPTBot, ClaudeBot, CCBot) collect content for models, AI search crawlers (OAI-SearchBot, Claude-SearchBot, PerplexityBot) find pages to cite, and user-triggered fetchers (ChatGPT-User, Perplexity-User) open a page because someone asked. The common mistake is blocking search when you only meant to opt out of training.

The repo has:

• data/: 27 crawlers as CSV and JSON, with category, training use, effect of blocking and the operator's docs linked on every row
• policies/: three ready-made blocks (allow search and block training, block all AI, allow all)
• A checker in JavaScript (Node 18+) and Python 3.8+, no dependencies, that follows RFC 9309 matching: the most specific user-agent group wins, the longest rule wins, Allow wins ties, and * and $ wildcards work

node js/robots-ai-check.mjs https://www.example.com
python python/robots_ai_check.py ./robots.txt /blog/ --json

2. awesome-geo

https://github.com/313Automations/awesome-geo

A curated list for Generative Engine Optimization that sticks to primary sources: Google's AI features guidance, Bing's webmaster guidelines (which now have a GEO section), OpenAI's publisher FAQ, each major AI crawler's own documentation, RFC 9309, llms.txt, Schema.org, IndexNow and the original GEO paper. CC0, and pull requests are welcome.

3. llms-txt-examples

https://github.com/313Automations/llms-txt-examples

Ten fictional businesses (dental clinic, law firm, plumber, restaurant, SaaS, online store, real estate, accountant, marketing agency, nonprofit), one llms.txt each, plus validate_llms_txt.py, a zero-dependency validator for files or live URLs:

python validate_llms_txt.py examples/*.llms.txt
python validate_llms_txt.py https://www.example.com

A note we also make in the README: llms.txt is a proposal. It is cheap to add and can help agents navigate a site, but no major AI search engine has confirmed it as a ranking or citation signal.

4. smb-automation-recipes

https://github.com/313Automations/smb-automation-recipes

30 recipes across sales, marketing, customer support, operations, finance, HR, e-commerce, real estate and clinics. Each one has the trigger and steps, build notes for Zapier, Make and n8n (which built-in tools to reach for: Paths, Routers, Wait nodes, error handlers), a test checklist and the watch-outs that usually bite in production, like duplicate records, SMS consent and keeping a person in the loop for AI output.

Licenses

Code is MIT. Data and text are CC-BY-4.0, and the awesome list is CC0.

All repos: https://github.com/313Automations

Issues and pull requests are welcome. What should we open-source next?