Enterprise search

Search across all your connected knowledge sources with permission-aware, hybrid semantic and keyword retrieval.

Hadiq.io Enterprise Search lets you find anything across your organization’s knowledge in seconds — from engineering docs to sales call notes. Results are permission-aware by default: each user sees only the content they are authorized to access in the original source system.

How search works

Hadiq.io uses a hybrid retrieval approach that combines semantic search (meaning-based) with keyword search (exact-match). Both signals are merged and re-ranked using deep learning models, so you get relevant results whether you phrase your query precisely or conversationally.

Under the hood, Hadiq.io:

1. Continuously indexes content from all connected sources in the background
2. Embeds documents using in-house deep learning models
3. Runs multi-pass indexing and contextual retrieval at query time
4. Merges semantic and keyword scores and re-ranks results
5. Applies your access control lists before returning any result

Because content is pre-indexed, searches return near-instantly — even across dozens of sources simultaneously.

Permission-aware results

Automatic permission syncing is available in Hadiq.io Enterprise. It syncs access control lists from Confluence, Jira, GitHub, Google Drive, Gmail, Slack, Salesforce, and SharePoint.

Hadiq.io syncs access control lists from your source systems automatically. When you run a search, Hadiq.io filters results to only include documents you are authorized to view in the original tool — no separate permission layer to configure or maintain.

Two users asking the same question can receive different results because each user’s permissions determine what Hadiq.io is allowed to surface. Your existing security policies carry over to AI search automatically.

Search across all your sources

Once your admin connects your organization’s tools, a single query reaches all of them at once. Hadiq.io currently supports 50+ connectors including:

• Confluence, Notion, Google Drive, SharePoint
• Slack, Gmail, Microsoft Teams
• GitHub, GitLab, Linear, Jira
• Salesforce, Zendesk, HubSpot
• And many more

See all connectors →

Search UI vs. Chat UI

Onyx provides two interfaces that serve different needs:

Search

The Search UI returns a ranked list of document results with highlighted excerpts and source citations. Use it when you want to browse and compare multiple sources, or when you know what you’re looking for and want direct links back to the source.

Best for:

• Finding a specific document or page
• Reviewing multiple sources on a topic
• Auditing what’s available in a knowledge base

Chat

The Chat UI synthesizes an AI-generated answer using your connected knowledge as context. Use it when you want a direct answer rather than a list of documents to read through.

Best for:

• Getting a summarized answer to a question
• Asking follow-up questions in context
• Combining knowledge from multiple sources in a single response

Both interfaces respect the same permissions and use the same underlying index.

Configure what gets indexed

Admins can control exactly what Onyx indexes from each connected source:

• Specific folders, channels, workspaces, or repositories
• Custom refresh frequencies (how often Onyx re-syncs content)
• Embedding model selection per connector

Go to Admin → Connectors to manage connector settings.

Learn more about connectors →