The setup involves cloning the repository, configuring a `.env` file, and running a startup script that utilizes Docker Compose to manage services. Sherlog Canvas uses MCP servers for data source integration, allowing users to configure connections to various data sources like Grafana, PostgreSQL, and Prometheus. The system's architecture includes a FastAPI server, dependency tracking for reactivity, and AI orchestration for query planning. Configuration is managed through environment variables, and the platform supports both SQLite and PostgreSQL databases.
Key takeaways:
- Sherlog Canvas is a reactive Jupyter notebook-like interface designed for software engineering investigation tasks, integrating advanced AI capabilities for debugging, log analysis, metric analysis, and database queries.
- The platform supports multiple cell types, including markdown, Python, GitHub, log analysis, filesystem access, summarization, and investigation reports, allowing for diverse data interactions and analyses.
- Sherlog Canvas operates with a dependency graph that automatically updates dependent cells when changes occur, ensuring a reactive and up-to-date investigation environment.
- Configuration and setup are managed through environment variables, with options for Docker-based deployment and integration with various data sources via MCP servers.