Connectors & openness

AI connectors to let your knowledge circulate.

Your knowledge bases can take content in and feed uses out. The AI connectors available and the integrations to plan for are set out here.

What can be connected

AI connectors, from sources to applications.

8 possible connections

Documents & files

PDFs, office documents and text. Which formats are available depends on the options in your plan.

Import into the platformLet us talk about your connection

Models and AI connectors

Connectors to the providers on offer, including OpenAI, Anthropic, Google, Mistral and local options.

Depending on configurationLet us talk about your connection
An open architecture

AI connectors are a means.
Your use case is the starting point.

A tool with an API may open possibilities. Feasibility depends on its interfaces, its permissions and the data to be sent.

We start from the work to be done, then the corpus it needs, before defining the connection and the controls required.

DocumentsWebsitesTrade knowledgeConnected tools
Your secured knowledge bases

Chosen sources · Defined access · Reusable knowledge

Inside Narratheque

AI for teams, assistants, agents, chatbot

In your tools

Connected applications and automations

The model adapts to the use. Your knowledge stays organised.
Scoping an integration

Four questions before laying AI connectors.

01

Which data?

Sources, formats, volume and how often it is updated.

02

Which rights?

Users, reachable scopes and sensitive data.

03

Which action?

Searching, drafting, or writing into a tool.

04

Which controls?

Approval, monitoring, cost and error handling.

Frequently asked questions

What you want to know about AI connectors.

Can I connect any piece of software?

Every tool has to be looked at. An API or an export may allow an integration, but it does not guarantee that a connector already exists or that every action is possible.

Are bespoke integrations included in the subscription?

Their scope, their cost and their upkeep are defined separately. The platform’s plans should not be confused with an integration engagement.

Can my base feed an external tool?

That is a possibility to scope, given the interfaces available. You have to check what is sent, the permissions, and what the receiving tool does with the data.

Let us start simply

Which AI connectors do you want to put in place?

Describe the flow you want: where the data comes from, where it should be useful, and which AI connectors to plan for.

RDV démo

Besoin d'aide ?