Paige — The AI Knowledge API for Your Business

THE PROBLEM

Your team is the search engine.

Someone asks a question. Someone else opens three tabs, two spreadsheets, and a shared drive folder.

Your data lives in silos.

Contracts in Google Drive. Client records in a spreadsheet. Website copy in a CMS. Nothing talks to anything.

You can't give an AI your whole business.

Generic chatbots don't know your data. Building a custom pipeline takes months.

INTRODUCING PAIGE

The brilliant library clerk your business never had.

Imagine the most knowledgeable colleague in the world. She has read every document your company has ever produced. She knows every spreadsheet, every contract, every web page, every API your systems expose.

Ask her anything. She doesn't search — she already knows where to look. She pulls the answer in under 500 milliseconds and tells you exactly where she found it.

That's Paige.

Unlike a search bar, Paige understands the intent behind a question. Unlike a generic chatbot, she knows your business specifically. And unlike building your own pipeline, she's ready in hours — not months.

"Paige isn't a chatbot. It's the intelligence layer that sits between your data and everything that needs to understand it."

HOW IT WORKS

Data in. Answers out.

Input Sources

PDF & Word Documents

Excel & CSV Spreadsheets

Websites & Web Pages

Live APIs & Data Feeds

Google Drive / Cloud Storage

PAIGE

INGEST

UNDERSTAND

ANSWER

Output Consumers

JSON API

MCP Server

Chatbot / Widget

AI Agent Pipeline

< 500ms

Standard query response time

5 source types

Documents, APIs, web, structured data

1 endpoint

One API call. Any question. Any source.

SEE IT IN ACTION

Four queries. Four completely different routes.

Every question that hits Paige gets classified, routed and answered differently. Here's what happens under the hood.

roundabouts

Ambiguous / Keyword

Query Analyser → Classify as ambiguous → Hybrid search (tenant-scoped) → Best-match result

{
  "classification": "ambiguous",
  "confidence": 0.42,
  "answer": "Roundabouts Ltd — supplier onboarding docs...",
  "source": "documents/suppliers/roundabouts-ltd.pdf",
  "flags": ["low_confidence", "ambiguous_query"]
}

Vague in, best guess out — with honest confidence scoring so your app knows when to ask a follow-up.

Which way do you go round a roundabout?

General Knowledge

Query Analyser → Classify as general → LLM knowledge (no data lookup)

{
  "classification": "general_knowledge",
  "confidence": 0.97,
  "answer": "In the UK, you travel clockwise...",
  "source": "llm_knowledge",
  "data_accessed": false
}

Common questions answered instantly without touching your data — fast, cheap, zero retrieval cost.

What are Duplantis's latest results?

Ephemeral / Live Data

Query Analyser → Ephemeral flag set → Live API connector → Real-time result

{
  "classification": "ephemeral",
  "confidence": 0.91,
  "answer": "Mondo Duplantis cleared 6.26m at...",
  "source": "connector/sports_api",
  "cached": false,
  "retrieved_at": "2026-03-09T14:32:00Z"
}

Time-sensitive answers from live APIs, never stale cache — because last week's data isn't good enough.

What are the contacts for Acme Groundworks?

Structured / Entity Lookup

Query Analyser → Entity: "Acme Groundworks" → CRM connector (direct lookup)

{
  "classification": "structured_lookup",
  "confidence": 0.99,
  "answer": {
    "company": "Acme Groundworks",
    "contact": "Sarah Chen",
    "email": "s.chen@acmegroundworks.co.uk",
    "phone": "+44 117 496 0023"
  },
  "source": "connector/crm/contacts"
}

Named entity detected, direct structured lookup — exact answer with source citation, zero noise.

FOR DEVELOPERS & TECHNICAL TEAMS

One endpoint. Infinite applications.

Paige exposes a single unified REST API. Every response — regardless of source, query type, or connector — returns the same structured JSON envelope. Drop it into any application, agent framework, or orchestration layer.

POST /api/query

{
  "project_id": "proj_abc123",
  "query": "What are the payment terms in the Acme contract?"
}

→ Response

{
  "answer": "Net 30 from invoice date, with a 2% early payment discount.",
  "confidence": 0.94,
  "sources": [
    {
      "connector": "contracts-drive",
      "document": "Acme_MSA_2024.pdf",
      "chunk": "Payment terms: Net 30...",
      "page": 4
    }
  ],
  "query_type": "factual_retrieval",
  "response_time_ms": 312
}

MCP Compatible

Paige works natively with MCP (Model Context Protocol). Drop the Paige connector into Claude Desktop, Cursor, or OpenClaw and your AI agent can query your entire business knowledge base instantly.

OpenClaw Ready

Paige is designed to work as an intelligence layer inside multi-agent systems. Gumbo, Claudia, and any other agent in your stack can call Paige for grounded, source-cited answers.

Framework Agnostic

No SDK required. If you can make an HTTP request, you can use Paige. Works with LangChain, LlamaIndex, n8n, Make, Zapier, or any custom stack.

COMMERCIAL MODEL

Simple, predictable pricing.

Standard

Contact for pricing

Perfect for single-project deployments

Up to 3 connectors
Document, web, and structured data sources
Single-step query execution (< 500ms)
Unified JSON response envelope
Standard support

Thinking

Contact for pricing

For teams that need multi-step intelligence

Unlimited connectors
Live API connector with chaining
Multi-step query planning & execution
Parallel tool execution
Full reasoning trace in Query Lab
Priority support & onboarding

Paige is currently in early access. Pricing is agreed per tenant based on usage volume and connector complexity. Book a call to discuss your specific use case.

SECURITY & COMPLIANCE

Built for businesses that can't afford to get this wrong.

Data Isolation

Every tenant operates in a fully isolated namespace. Your data cannot be seen, queried, or accessed by any other tenant. Namespace enforcement is applied at every layer of the stack — ingestion, index, and query.

Encryption

All data is encrypted at rest and in transit. API keys are hashed and never stored in plain text. TLS 1.3 enforced on all endpoints.

SOC 2-Aligned Infrastructure

Paige is built on AWS managed services with SOC 2-aligned configurations. No data is stored in regions outside your agreed jurisdiction.

Audit Trail

Every query, ingestion event, and API call is logged with full metadata. Know exactly who asked what, when, and what source it came from.

TLS 1.3 | AWS Managed | Hashed API Keys | Audit Logging | Data Residency Options

Your business knows the answer.
Paige finds it.

Your team is the search engine.

Your data lives in silos.

You can't give an AI your whole business.

The brilliant library clerk your business never had.

Data in. Answers out.

Input Sources

Output Consumers

Four queries. Four completely different routes.

One endpoint. Infinite applications.

MCP Compatible

OpenClaw Ready

Framework Agnostic

Simple, predictable pricing.

Standard

Thinking

Built for businesses that can't afford to get this wrong.

Data Isolation

Encryption

SOC 2-Aligned Infrastructure

Audit Trail

Ready to give your business a brain?

Your business knows the answer.Paige finds it.

Your team is the search engine.

Your data lives in silos.

You can't give an AI your whole business.

The brilliant library clerk your business never had.

Data in. Answers out.

Input Sources

Output Consumers

Four queries. Four completely different routes.

One endpoint. Infinite applications.

MCP Compatible

OpenClaw Ready

Framework Agnostic

Simple, predictable pricing.

Standard

Thinking

Built for businesses that can't afford to get this wrong.

Data Isolation

Encryption

SOC 2-Aligned Infrastructure

Audit Trail

Ready to give your business a brain?

Your business knows the answer.
Paige finds it.