← ALL AI EMPLOYEES

Knowledge · Search · MODULE X

Knowledge Hub

Your company knowledge is everywhere. Your team needs it in one place.

Module X ingests your Google Drive, Notion, Confluence, SharePoint, and uploaded PDFs into a searchable index. Ask in plain English (or BM), get answers grounded in your documents — with citations. Powers Module A WhatsApp Receptionist, Module V Voice Receptionist, and your internal team.

6 sources

Drive · Notion · Confluence · SharePoint · PDF · web

Cited

Every answer links to its source doc

<2s

Search query to ranked answer

WHAT THIS FIXES

The exact problems Knowledge Hub is built around.

  • Your SOPs live in Notion, your contracts in Drive, your FAQs in Confluence — the answer your team needs is somewhere, but no one can find it fast.
  • New hires take 3-4 weeks to ramp up because they keep asking "where's X?" — the institutional knowledge is invisible.
  • Customer-facing teams give different answers to the same question — knowledge is fragmented across channels.
  • Compliance audits demand "show me the SOP that covers X" — you spend 2 days finding it.

HOW IT WORKS

The pipeline, end to end.

  1. 1

    Source connectors

    OAuth into Drive, Notion, Confluence, SharePoint. We index documents you authorise — never read what we shouldn't.

  2. 2

    Chunking + embeddings

    Documents split into semantic chunks. Each chunk embedded with openai/text-embedding-3-small (1536 dims). Stored in pgvector with tenant isolation.

  3. 3

    Ask in plain English

    Search bar or API call. Query embedded, top-K chunks retrieved, sentence-rewrite for context.

  4. 4

    Grounded answer

    LLM (Claude Sonnet 4.6) reads chunks + writes answer. Citations point to original docs.

  5. 5

    Embeds into other modules

    Module A (WhatsApp) and Module V (Voice) read from Knowledge Hub when answering customers. Your KB powers your AI workforce.

WHAT'S INCLUDED

Every capability shipped, named.

  • Google Drive, Notion, Confluence, SharePoint, PDF upload connectors
  • Semantic search (embedding-based) — not keyword
  • Source citations on every answer
  • BM + English + Mandarin queries
  • PDPA-compliant — RLS-enforced tenant isolation
  • Real-time index sync (Drive Notion changes within 60s)
  • Per-doc permissions respect (won't leak across teams)
  • Public API for embedding in your own apps
  • Powers Module A + Module V grounding
  • Audit log of every query (compliance-friendly)

PRICING

Plans that scale with use, not seats.

All prices in MYR, exclusive of SST. 14-day refund on monthly, pro-rated refund on annual. PDPA-aligned. Cancel anytime.

Solo

RM 999/mo

10k documents · 3 sources · 10 users

Start trial

MOST POPULAR

Team

RM 2,999/mo

100k documents · all sources · 50 users · API access

Start trial

Enterprise

RM 7,999/mo+

Unlimited · custom connectors · SSO · dedicated index

Talk to sales

WHAT YOU RECEIVE

The artefacts that land in your hands.

Search results UI

Top-3 answers ranked by semantic similarity, each with a snippet + source link. Click to open the original doc.

API endpoint

POST /search returns ranked chunks + LLM-grounded summary + citations. Drop into your own app via REST or MCP.

Q&A history

Every query logged with timestamp, user, retrieved chunks, generated answer. Filterable by user/topic. Compliance-friendly.

FREQUENTLY ASKED

The questions buyers ask first.

  • How does this differ from ChatGPT?

    ChatGPT answers from its training data (public web). Module X answers from YOUR documents only. No hallucination — every answer cites your specific doc.
  • Does it respect per-document permissions?

    Yes. Module X reads the same permissions Drive/Notion/SharePoint enforce. Users only see results from docs they're already authorised to read.
  • BM queries work?

    Yes — embeddings are multilingual. Search in BM, get results from BM/EN/中文 docs ranked by semantic similarity, not language match.
  • How fresh is the index?

    Drive/Notion/Confluence webhooks update within 60s of a doc change. SharePoint via scheduled pull every 5 minutes. PDFs immediately on upload.
  • API for embedding in our apps?

    Yes — Team tier includes REST API access. Also exposed via MCP (Model Context Protocol) for Claude desktop / Cursor / Cline integrations.
  • PDPA compliance?

    Documents stored in MY/SG region (Supabase ap-southeast-1). RLS-enforced tenant isolation. Per-tenant retention configurable. Right-to-erasure 1-click via /admin.

Try Knowledge Hub on your real business.

14-day trial. No credit card. PDPA-aligned. Built for Malaysian SMEs by a Malaysian team in KL. Reginald and ONSET team reply within one business hour, Mon–Fri 9am–6pm MYT.

  • PDPA-aligned
  • Cancel anytime
  • 14-day refund
  • 1 business hour reply