Community-built production-ready tools for Strands Agents SDK. Build powerful AI agents with seamless integrations for speech processing, CRM operations, and team notifications.
This is a meta-package that bundles three Strands community tools for convenience. For new projects following Strands conventions, consider using individual packages:
# Individual packages (recommended for new projects)
pip install strands-deepgram # Speech processing
pip install strands-hubspot # CRM operations
pip install strands-teams # Teams notifications
# Or install all three via meta-package
pip install strands-tools-communityIndividual Package Links:
- 🎤 strands-deepgram - Speech & audio processing
- 🏢 strands-hubspot - HubSpot CRM operations
- 📢 strands-teams - Microsoft Teams notifications
This package provides three production-ready tools that extend the capabilities of Strands agents:
| Tool | Description | Key Features |
|---|---|---|
| Deepgram | Speech & Audio Processing | Speech-to-text, text-to-speech, audio intelligence, 30+ languages, speaker diarization |
| HubSpot | CRM Operations | Universal CRM access, smart search, CRUD operations, associations, batch processing |
| Microsoft Teams | Notifications & Alerts | Adaptive cards, pre-built templates, custom notifications, rich formatting |
# Install the community tools
pip install strands-tools-community
# Install Strands SDK with your preferred model provider
pip install 'strands-agents[anthropic]' # For Anthropic Claude (recommended)
# OR
pip install 'strands-agents[openai]' # For OpenAI GPT
# OR
pip install 'strands-agents[bedrock]' # For AWS Bedrockfrom strands import Agent
from strands_tools_community import deepgram, hubspot, teams
# Create an agent with all tools
agent = Agent(tools=[deepgram, hubspot, teams])
# Use natural language to interact
agent("transcribe this audio file: recording.mp3")
agent("search for contacts in HubSpot with email containing '@example.com'")
agent("send a Teams notification about new leads")Powerful speech processing capabilities powered by Deepgram's API.
Key Features:
- Speech-to-Text: Transcribe audio with 30+ language support
- Text-to-Speech: Generate natural-sounding speech in multiple voices
- Audio Intelligence: Sentiment analysis, topic detection, intent recognition
- Speaker Diarization: Identify and separate different speakers
- Multi-format Support: WAV, MP3, M4A, FLAC, and more
Usage Example:
from strands import Agent
from strands_tools_community import deepgram
agent = Agent(tools=[deepgram])
# Transcribe audio with speaker identification
agent("transcribe audio from https://example.com/call.mp3 in Turkish with speaker diarization")
# Text-to-speech
agent("convert this text to speech and save as output.mp3: Hello world")
# Audio intelligence
agent("analyze sentiment and topics in recording.wav")Configuration:
DEEPGRAM_API_KEY=your_deepgram_api_key # Required
DEEPGRAM_DEFAULT_MODEL=nova-3 # Optional
DEEPGRAM_DEFAULT_LANGUAGE=en # OptionalGet your API key at: console.deepgram.com
Complete CRM integration for managing contacts, deals, companies, and more.
Key Features:
- Universal CRM Access: Works with ANY HubSpot object type
- Smart Search: Advanced filtering with property-based queries
- CRUD Operations: Create, read, update, and delete records
- Property Discovery: Automatic field detection and validation
- Association Management: Link related objects (contacts, deals, companies)
- Rich Console Output: Beautiful table displays with Rich library
Usage Example:
from strands import Agent
from strands_tools_community import hubspot
agent = Agent(tools=[hubspot])
# Search contacts
agent("find all contacts created in the last 30 days")
# Create a deal
agent("create a deal called 'Acme Corp Q4' with amount 50000")
# Update records
agent("update contact 12345 with lifecycle stage 'customer'")
# Get company details
agent("get company information for ID 67890")Configuration:
HUBSPOT_API_KEY=your_hubspot_api_key # Required
HUBSPOT_DEFAULT_LIMIT=100 # OptionalGet your API key at: app.hubspot.com/private-apps
Send rich, interactive notifications to Microsoft Teams channels.
Key Features:
- Adaptive Cards: Rich, interactive message cards
- Pre-built Templates: Notifications, approvals, status updates
- Custom Cards: Full adaptive card schema support
- Action Buttons: Add interactive elements to messages
- Rich Formatting: Markdown support, images, and media
Usage Example:
from strands import Agent
from strands_tools_community import teams
agent = Agent(tools=[teams])
# Simple notification
agent("send a Teams message: New lead from Acme Corp")
# Use pre-built templates
agent("send an approval request to Teams for the Q4 budget")
# Status updates
agent("send a status update to Teams: website redesign is 75% complete")Configuration:
TEAMS_WEBHOOK_URL=your_webhook_url # Optional - can be provided per callSetup webhook: Teams Channel → Connectors → Incoming Webhook
from strands import Agent
from strands_tools_community import deepgram, hubspot, teams
agent = Agent(tools=[deepgram, hubspot, teams])
# Automated call processing
agent("""
1. Transcribe the call recording at: call_recording.mp3
2. Search HubSpot for the contact with phone number from the call
3. Create a call activity in HubSpot with the transcript
4. Send a Teams notification about the call summary
""")# Qualify and notify about new leads
agent("""
Search HubSpot for all leads with:
- Lifecycle stage: Marketing Qualified Lead
- Created in the last 7 days
- Company size: 50+ employees
Then send a summary to Teams with the count and top 5 leads
""")# Automated daily reporting
agent("""
1. Get all deals closed today from HubSpot
2. Get all new contacts created today
3. Calculate total revenue
4. Send a formatted digest to Teams with the results
""")All tools support environment variables for API keys and default settings. You can also pass configuration directly when calling tools through the agent.
# Deepgram Configuration
DEEPGRAM_API_KEY=your_deepgram_api_key
DEEPGRAM_DEFAULT_MODEL=nova-3
DEEPGRAM_DEFAULT_LANGUAGE=en
# HubSpot Configuration
HUBSPOT_API_KEY=your_hubspot_api_key
HUBSPOT_DEFAULT_LIMIT=100
# Microsoft Teams Configuration
TEAMS_WEBHOOK_URL=your_teams_webhook_url- Deepgram: console.deepgram.com
- HubSpot: app.hubspot.com/private-apps
- Teams Webhook: Teams Channel → Connectors → Incoming Webhook
For more information about the Strands Agent SDK:
Contributions are welcome! Please feel free to submit a Pull Request.
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
This project is licensed under the MIT License - see the LICENSE file for details.
- Built for the Strands Agent SDK
- Powered by Deepgram, HubSpot, and Microsoft Teams
- Issues: GitHub Issues
- Discussions: GitHub Discussions
Made with ❤️ by the community