Cross-Document Attribute Filtering & Aggregate Search

Currently, Chatling can retrieve attributes from individual documents (e.g., “What is the launch window for XYZ”). However, it cannot process queries that require scanning across all documents to filter by attribute values (e.g., “List all products launching in 2026”).

For organisations managing large product catalogues or structured datasets, this limitation prevents the chatbot from delivering full value. Attributes (such as Launch Window, SAP fields, Brand, etc.) are often stored in HTML tables across many pages, and being able to query them globally would transform the platform’s utility.

Proposed Functionality

• Support cross-document filtering: search across all uploaded knowledge base documents for specific attribute fields and values.

• Return a list of matching entries, not just single responses.

• Enable multi-attribute queries, e.g., “List all products where Launch Window = 2026 and Category = XYZ”

• Provide tabular responses (rows/columns of Name/Value pairs) for aggregate results.

Example Use Cases

• Product catalogues (list all SKUs by launch date, category, or specification).

• Knowledge bases with structured data (compliance docs, training modules, HR policies, etc.).

• Any scenario where attributes are spread across many documents but need to be queried as a set.

Impact
This feature would significantly improve Chatling’s value for enterprise users who rely on structured, attribute-rich datasets. It eliminates the need for manually consolidating thousands of pages into one document and makes Q&A far more powerful and efficient.

Workarounds Tried

• Creating a single consolidated document with key attributes.

  • Works in part, but is not scalable for large, frequently updated datasets.

Request
Please prioritise this feature for development. It would enable Chatling to handle attribute-based queries at scale, unlock new enterprise use cases, and improve adoption for data-heavy industries.