Torale DocsTechnical Documentation
Grounded search monitoring platform
Skip to content
Overview
Torale executes scheduled web searches, evaluates results against trigger conditions, and stores execution history. Grounded search combines Google Search with LLM evaluation for source-backed monitoring.
Example
python
from torale import ToraleClient
client = ToraleClient()
# Create a monitoring task
task = client.tasks.create(
search_query="When is the iPhone 17 being released?",
condition_description="Apple has announced a specific release date",
schedule="0 9 * * *" # Daily at 9 AM
)
# Check results
executions = client.tasks.get_executions(task.id)
if executions[0].condition_met:
print(executions[0].result["answer"])Architecture
Core Components:
- Grounded Search: Google Search API via Gemini LLM with source attribution
- Temporal Workflows: Cron-based scheduling with automatic retries and state management
- State Tracking: Detects changes between executions to prevent duplicate notifications
- Condition Evaluation: LLM-based determination of trigger condition status
Interfaces
- Web Dashboard - Browser-based task management
- CLI - Command-line interface
- Python SDK - Programmatic access
- API - REST API reference
- Self-Hosted - Deploy on your infrastructure