Skip to content

babelwrap/babelwrap-examples

Repository files navigation

Babelwrap Examples

Official example projects for BabelWrap — the web, as an API, for your agents.

BabelWrap lets AI agents interact with any website through a simple API using natural language instead of DOM selectors. These examples show how to build agents that use BabelWrap to navigate the web, fill forms, shop, research, handle authentication, and automate complex workflows.

Examples

# Example Framework Website What It Does
1 hello-babelwrap Python SDK books.toscrape.com Quick intro — navigate, click, extract (no LLM key needed)
2 map-and-discover Python SDK books.toscrape.com Site mapping — map a site, use generated tools (no LLM key needed)
3 claude-shopping-agent Anthropic SDK saucedemo.com Claude agent logs in, shops, checks out + screenshots
4 openai-web-researcher OpenAI SDK Wikipedia + real sites GPT-4o agent researches topics with scroll + extract
5 agno-web-automation Agno + Claude the-internet.herokuapp.com Auth, hover, keyboard, checkboxes, file upload — every interaction
6 agno-deal-finder Agno + Claude saucedemo.com Site mapping + batch actions for efficient deal finding
7 multi-site-researcher Anthropic SDK multiple sites Cookie auth persistence + screenshots + mapping

Feature Coverage

Feature Ex1 Ex2 Ex3 Ex4 Ex5 Ex6 Ex7
Navigate, click, extract X X X X X X
Fill forms + submit X X X
Site mapping + tools X X X
Screenshot X X
Scroll pagination X X
Hover X
Keyboard press X
File upload X
Batch actions X
Cookie auth persistence X

Getting Started

1. Get API Keys

2. Clone and Configure

git clone https://github.com/babelwrap/babelwrap-examples.git
cd babelwrap-examples
cp .env.example .env
# Edit .env and add your API keys

3. Run Any Example

Each example has its own README. Start with Example 1 for basics, then jump to any agent example:

cd 05-agno-web-automation
pip install -r requirements.txt
python web_automation.py

How the Examples Are Organized

Examples 1-2 teach the BabelWrap SDK fundamentals — sessions, navigation, data extraction, and site mapping. No LLM API key needed.

Examples 3-7 are the main event — AI agents using BabelWrap as their web interaction tool. Each showcases different capabilities on different websites:

  • Ex 3: Full e-commerce checkout with visual proof (screenshots)
  • Ex 4: Research across real websites with scroll+extract pagination
  • Ex 5: Every interaction type — auth, hover, keyboard, upload, checkboxes
  • Ex 6: Site mapping + batch operations for efficiency
  • Ex 7: Authentication persistence with cookies + multi-site research

Resources

About

Example projects showing how to use BabelWrap — AI agents interacting with websites via natural language.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages