README.md

metadata

title: First Agent Template
emoji: ⚡
colorFrom: pink
colorTo: yellow
sdk: gradio
sdk_version: 5.20.1
app_file: app.py
pinned: false
tags:
  - smolagents
  - agent
  - smolagent
  - tool
  - agent-course

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

AI Assistant Agent with Useful Tools

This Space provides an AI assistant agent built with the smolagents framework that can perform several useful tasks through specialized tools. The agent uses the Qwen2.5-Coder-32B-Instruct model to understand user requests and choose the appropriate tool to fulfill them.

What This Agent Does

This agent serves as an intelligent assistant that can:

• Analyze text and provide detailed statistics
• Get current time in different timezones around the world
• Perform semantic search on text to find relevant information
• Provide documentation about its own capabilities

The agent uses a Gradio interface for easy interaction through natural language. Simply type your request, and the agent will handle the rest!

Available Tools

1. Text Analyzer

This tool analyzes text and provides comprehensive statistics including:

• Word count
• Character count
• Unique words count
• Average word length
• Most common words (top 5)

2. Current Time in Timezone

Gets the current local time for any timezone in the world using the pytz library. The tool accepts standard timezone format names like:

• 'America/New_York'
• 'Europe/London'
• 'Asia/Tokyo'
• 'UTC'

3. Semantic Search

This tool performs semantic search on a corpus of text to find the most relevant sections matching a query. It uses:

• Bag-of-words vectorization for text representation
• Cosine similarity for relevance scoring
• Smart text chunking for better results

Note: While designed to also work with PDF documents, the current UI doesn't support file uploads. The semantic search works with text provided directly in the conversation.

4. Tools Documentation

A special tool that lists all available tools and provides usage examples for each. Access this by asking about the agent's capabilities.

Example Queries to Try

Text Analysis

• "Analyze this text: The quick brown fox jumps over the lazy dog."
• "Give me statistics about this paragraph: Lorem ipsum dolor sit amet, consectetur adipiscing elit."

Timezone Queries

• "What time is it in Tokyo?"
• "Get the current time in America/New_York"
• "Tell me the time in UTC"

Semantic Search

• "Search for 'climate change' in this text: Global warming is the long-term heating of Earth's surface observed since the pre-industrial period due to human activities, primarily fossil fuel burning, which increases heat-trapping greenhouse gas levels in Earth's atmosphere."
• "Find information about 'artificial intelligence' in this text: [paste your text here]"

Getting Help

• "What tools do you have?"
• "Show me how to use this agent"
• "Help"

Requirements and Dependencies

This agent requires:

• smolagents framework
• pytz for timezone functionality
• numpy for semantic search vector operations
• PyPDF2 (optional, for PDF processing if enabled)

Usage Tips

1. Be specific in your requests
2. For semantic search, provide substantial text to search within
3. For timezone queries, use standard timezone names

Limitations

• The current UI does not support file uploads for PDF processing
• Semantic search uses a simplified algorithm and works best with moderate amounts of text
• The agent has a maximum context length, so very large inputs may be truncated

Future Development

Plans for future enhancements include:

• Adding file upload capabilities
• Implementing more sophisticated semantic search algorithms
• Adding additional tools like web scraping (when environment permits)
• Improving the PDF handling capabilities

---

Feel free to contribute to this project by providing feedback or suggesting new tools that would be useful!