Utilized alongside PyTorch for NLP tasks within the GPU-based document extraction pipeline.

A temporary unified view created to enhance Arctic query handling over split tables, based on schema relationships, to improve answer accuracy.

Process of creating temporary unified views from related tables to improve Arctic query execution.

Transient views created during consolidation are used as input schema by Arctic for SQL generation.

question_router.py uses translate_duckdb_to_pg to convert DuckDB-style SQL to PostgreSQL-compatible SQL before execution.

The Frontend Components include the TrendChart.svelte UI component which renders SVG line charts with grid. The TrendChart.svelte UI component is part of the Full Findings Visualization Layer and provides SVG line charts with grid. The TrendChart.svelte UI component is part of the Full Findings Visualization Layer and provides SVG line charts with grid.

The Frontend uses TypeScript as a core technology for type safety.

The Data Discovery feature includes UI components. The UI components depend on the Discovery backend service to function correctly.

UI components for discovery and consolidation review have been developed with responsive design. Frontend UI Components implement the user interface elements for editing and displaying the scope as the Project Goal.

The UI Flow allows the User to enter Danish questions and interact with the Data Discovery feature. The UI Flow is implemented using Frontend components such as DiscoveryFlow, DiscoveryLoading, ConsolidationReview, and ConsolidationCard. The Frontend DiscoveryFlow component is part of the UI Flow for the Data Discovery feature.

UI audit tools are used within the DataLens platform backend. The DataLens platform backend includes the ui-audit skill. The DataLens platform backend integrates the ui-audit skill.

The Unified /ask endpoint uses the Question router to process queries, replacing the SQL-only /query endpoint. The unified ask use case is realized by the Unified /ask endpoint integrating structured and semantic querying.

The Unified analysis page implements the unified question interface providing a combined SQL and semantic query experience. The Unified analysis page uses the Question router for classification badges and query routing. The unified analysis page is validated by the analysis.spec.ts frontend test case.

The unified ask capability is realized in Phase C, which provides a unified question interface combining SQL and semantic routing. The batch upload pipeline is integrated with the unified ask interface to provide query capabilities. DataLens provides the unified ask interface for natural language question routing over structured and semantic data.

The unified question interface routes structured queries to DuckDB SQL within DataLens. The unified question interface routes semantic search queries via Qdrant in DataLens. The Unified analysis page implements the unified question interface providing a combined SQL and semantic query experience. The unified ask use case is realized by the Unified /ask endpoint integrating structured and semantic querying. The unified ask capability is realized in Phase C, which provides a unified question interface combining SQL and semantic routing.

The unified question interface is implemented as part of the DataLens platform.

The consolidated view contains the unified schema for querying purposes.

The DataLens Platform supports a file upload feature for CSV, Excel, and PDF files. The File upload feature in the DataLens Platform uses python-multipart to handle multipart form data uploads. aiofiles is used in the File upload feature of the DataLens Platform to handle asynchronous file operations.

Stores vectors in Qdrant collection for project.

Represents a user of the system, external stakeholder with variable influence. The UI Flow allows the User to enter Danish questions and interact with the Data Discovery feature. The User is expected to provide the Anthropic API key User uses the Agent Chat interface to interact with the SVGV budget analysis system User asks Danish language budget queries to the DataLens SVGV Budget analysis system Organization physical table includes multiple User entities representing users belonging to an organization. The queries table includes a user_id column that associates each query with a user. The file_uploads table includes an uploaded_by field that links each file to the user who uploaded it. Each agent session is associated with a user via the user_id column in agent_sessions. User interacts with Project data via the API, querying and managing project-specific information. User interacts with IronClaw Gateway as part of the platform to ask Danish budget questions and receive clarified and executed results. User accesses the theo Backend as main API and service endpoint to interact with the platform for budget analysis. Frontend serves the User interface for querying and displaying budget analysis results in Danish. User interacts with OpenClaw as the agent orchestration platform to ask Danish budget questions. OpenClaw HTTP streaming serves real-time token streaming to the User interface for interactive experience. Playwright is used for end-to-end testing to validate User interactions and the platform's functionality. Organization physical table contains User physical table as members belonging to the organization. PostgreSQL database stores user information for authentication and project tracking. Query data entity references the User entity by user_id.