Search approved documents, business connections and embedded files with source evidence. Organise access and keep knowledge current.
Bring approved documents, transcripts, structured records and business-system read actions into the relevant topic.
Preserve document identities, versions, timestamps and parent-child relationships so evidence can be traced back.
Inspect the cited material and confirm that the source is suitable for the business decision.
Start with 50 preselected connectors, preconfigured by platform administrators for easy enablement, not a hard limit. Additional supported connectors can be configured as needed. Connections include Google Drive, Google Sheets, Google Calendar, Microsoft OneDrive, Microsoft Teams, Slack, Jira, Confluence, Notion, GitHub, HubSpot, Zendesk, Airtable, Dropbox, Google BigQuery and Databricks. Credentials, action review and topic approval are required.
Process supported files embedded in Office documents, PDFs and email files, where enabled. Keep child documents linked to the parent so users can understand where cited evidence came from. Supported formats, nesting and processing limits apply; access and lifecycle controls govern the document family.
Transcribe supported audio files and the audio tracks of video files, retaining timestamps for source references. Machine-generated transcription requires a compatible configured model; it does not provide general video-frame understanding.
Retain declared types, nested relationships and labelled paths in supported JSON records. Retrieve exact record fields alongside ordinary document evidence.
Set document routes and priorities within a topic, with explicit fallback and disclosure settings. Keep the current searchable version available while a replacement is processed.
Support 25+ document formats and 40+ storage connectors, alongside databases, websites and APIs. Supported media can use document imports, configured cloud imports and eligible conversation uploads. Attach up to ten files per question.
Combine hybrid search, query decomposition, reranking and grounding checks. Preserve exact phrases, named sections, definitions, table identities and page references, linking individual claims to their supporting evidence.
Bilingual ingestion and structural anchors support Malay, Chinese and Tamil content. An automatically built, bilingual topic wiki stays under administrator control.
GraphRAG combines graph and vector evidence in a ranked pool, with provenance on extracted facts, drift detection and verified deletion.
Select a focused topic mode using one authorised vector search and one grounded answer call for straightforward document questions. Flash topics skip dedicated decision tasks.
Availability depends on topic permissions, platform configuration, supported models and deployment. Confirm the required workflow with representative sources during evaluation.
Bring representative sources, a useful business question and the controls your organisation needs.
