Writes & summarizes
Drafts, syntheses, minutes and rewrites — in conversations you can organize into folders, with full Markdown rendering.
An AI assistant that acts across all your apps — without ever sending your data to someone else's cloud.
By: Kubuno
Listing updated: Aperçu
Official module, published and maintained by Kubuno
Assistant is Kubuno's cross-cutting intelligence: an assistant that, with your permission, acts on your apps through tools — listing your files, creating an event, launching a workflow — and helps you draft, summarize and rephrase in natural language.
The difference with mainstream assistants: Assistant runs on your infrastructure, with the models you choose — a local Ollama engine for total confidentiality, or the OpenAI, Anthropic or Google API if you decide so. The administrator can switch remote engines off entirely: no data leaves the server then. AI finally serving your data, not the other way around.
Drafts, syntheses, minutes and rewrites — in conversations you can organize into folders, with full Markdown rendering.
Build reusable assistant profiles — instructions, preferred model, suggestions — and share system agents with the whole instance.
List your files, create an event, launch a Flow workflow: Assistant acts through tools, and every sensitive action waits for your explicit confirmation.
Run an open-weight model on your own machine through Ollama: no data leaves, not even the prompt.
Prefer a large remote model? Plug in the OpenAI, Anthropic or Google API. The administrator can cut remote engines off with a single switch.
No training on your data, no profiling: your assistant stays strictly at home, with an admin-driven model allowlist.
Summarize reports, prepare replies and draft your writing in seconds.
Trigger your workflows and create your events in natural language, without leaving the conversation.
Healthcare, legal, public sector: benefit from AI without ever exposing confidential data to a third party.
No reviews yet — yours will be the first.