Help extend Goose's capabilities by creating a toolkit that allows it to read and process content from web links. This will enable our AI developer agent to access and analyze online information directly.
🎯 Objective
Develop a toolkit that enables Goose to open web links in a browser, extract content, and make it available for further processing or analysis.
🛠️ Implementation
Create a new class that derives from the Toolkit class
Implement tools as methods decorated with @tool
Provide a system() method for any necessary system instructions
✨ Key Features to Consider
Open web links in a headless browser (e.g., using Selenium or Playwright)
Extract text content from web pages
Handle different types of web content (articles, documentation, etc.)
Implement basic web scraping capabilities
Provide options for content summarization or specific data extraction
📋 Requirements
Familiarity with Python and web scraping libraries (e.g., Selenium, Playwright, BeautifulSoup)
Understanding of Goose's toolkit structure
Knowledge of HTML and basic web technologies
🚀 Getting Started
Fork the Goose repository (or goose-plugins if appropriate)
Add your toolkit class (e.g., class WebContentReader(Toolkit):)
Implement necessary tools for web content reading and processing
Update pyproject.toml to expose the new toolkit
Create a sample profile in profiles.yaml that uses the new toolkit
📚 Toolkit Integration
Add your Web Content Reader toolkit to the pyproject.toml:
web-reader-profile:
provider: openai # or any other provider
processor: gpt-4o
accelerator: gpt-4o-mini
moderator: passive
toolkits:
- name: developer
requires: {}
- name: web-content-reader
requires: {}
For detailed instructions on creating toolkits, refer to our documentation.
We welcome contributions of all levels. If you have any questions, feel free to ask in the comments!
🌐 Create a Web Content Reader Toolkit for Goose
Help extend Goose's capabilities by creating a toolkit that allows it to read and process content from web links. This will enable our AI developer agent to access and analyze online information directly.
🎯 Objective
Develop a toolkit that enables Goose to open web links in a browser, extract content, and make it available for further processing or analysis.
🛠️ Implementation
Toolkit
class@tool
system()
method for any necessary system instructions✨ Key Features to Consider
📋 Requirements
🚀 Getting Started
class WebContentReader(Toolkit):
)pyproject.toml
to expose the new toolkitprofiles.yaml
that uses the new toolkit📚 Toolkit Integration
pyproject.toml
:For detailed instructions on creating toolkits, refer to our documentation. We welcome contributions of all levels. If you have any questions, feel free to ask in the comments!