Requirements

To use the AI Assistant, you need an account and API key from one of the supported providers:

  • OpenAI - GPT-4o, GPT-4, o1, o3, gpt-5 and other models
  • Gemini (Google) - gemini-2.0-flash, gemini-1.5-pro and other models
  • Claude (Anthropic) - claude-opus-4, claude-sonnet-4 and other models
  • Grok (xAI) - grok-3, grok-2 and other models

The names listed are just examples of model families - the app fetches the full, up-to-date list from the provider itself once you enter a key.

Step by step - setting up AI

1. Open the settings

Click the gear icon (⚙) in the top-left corner → AI tab.

2. Select a provider

Choose one of the available providers from the “AI provider” list.

3. Enter the API key

Paste the API key obtained from the provider. The key is encrypted and is never displayed in full after it has been saved.

Where can I get an API key?

  • OpenAI: https://platform.openai.com/api-keys
  • Gemini: https://aistudio.google.com/apikey
  • Claude: https://console.anthropic.com/settings/keys
  • Grok: https://console.x.ai/

4. Select a model

After entering the API key, click the refresh button (↻) next to the model list - the app will fetch the current list of available models from the provider. Choose a model from the list.

Models marked (recommended) are proven, stable choices. Exactly one model per provider gets this label: gpt-4o (OpenAI), gemini-2.0-flash (Gemini), claude-opus-4-8 (Claude) and grok-3 (Grok). The other models on the list work the same way - they just don't have this label.

5. Save the configuration

Click “Save AI settings”. The app will test the connection to the model and display any warnings.

Task Manager: AI settings with a list of providers, an API key field and a list of models.
Settings → AI: provider, key and the list of models fetched from the provider.

Advanced settings

In the “AI settings” section, several options are grouped in the expandable Advanced subsection (click the heading to expand it):

  • Maximum response length - controls the length of the AI's responses (1024-16384 tokens). A higher value means longer responses, but higher costs. Default: 4096.
  • Save AI Assistant conversation history - the administrator can turn saving of conversation history in the database on or off. Enabled by default.
  • AI attachment summaries - AI creates short summaries of files to respond faster in tasks with many attachments. Set to Manual by default (summaries are only created after clicking Create summary next to the section name - no automatic API usage). The list of skipped/failed files is available under "Details".
  • Semantic search - search by meaning, not just by keywords. Requires an OpenAI or Gemini provider (enabled by default for these providers).
  • Allow mail to be used as an AI knowledge source - disabled by default. Once enabled, each user can point to their own mailbox or email thread as a source for a conversation with the Assistant (nobody, including the administrator, can see someone else's mail). The content of the indicated source is then sent to the external AI provider, and correspondence contains third-party data - hence the administrator's informed consent is required. Mail never goes into the knowledge base indexed in the background; turning the switch off also cuts off previously indicated mail sources.
  • Daily AI query limit from Automations - the condition "AI evaluates" and the action "write via AI" query the model on every matching event; the limit (200/day by default) protects your budget with the provider. Today's usage is shown next to the field. Entering 0 disables the limit. Once the limit is used up, AI-based recipes are skipped for the rest of the day with an entry in the history - the counter resets at midnight and everything starts again on its own.

AI knowledge base

The “AI Knowledge Base” section in AI settings (administrators only) lets you manage the Assistant’s automatic context.

The AI knowledge base belongs to the Pro plan. In the Standard plan, there's no such section nor a "Semantic search" toggle - the knowledge base isn't built, and the AI Assistant answers based on the conversation you're asking in and the sources indicated in its window.

Automatic knowledge base updates

The Automatic knowledge base updates switch (Yes / No) controls whether the knowledge base is refreshed automatically:

  • Yes (default) - the knowledge base updates automatically once a day in the background. Users can carry on using the app as normal.
  • No - auto-update is disabled. The administrator updates the base manually with the Update AI knowledge base button whenever needed. Recommended with a very large number of tasks or chats.

Manual update

The Update AI knowledge base button immediately processes all tasks and chats that do not yet have a summary or have been changed since the last update. It works regardless of the automatic update setting.

What is included in the knowledge base

  • Tasks and chats - summarised by the model (decisions, conclusions and agreements).
  • Notes and Files (from 2.2.273) - these are sent there in their original content, split into fragments. This lets the assistant find the answer in company documents by itself, even if no one points to them as a source - just ask "what's the fuel limit?". The model doesn't summarise documents. With OpenAI and Gemini, when semantic search is enabled, fragments are sent to the provider for vectors (billed separately); with Claude and Grok, indexing doesn't send anything to the provider. Indexing runs together with the knowledge base update, so it only works when AI is configured. From files, text formats are read: TXT, CSV, Markdown, DOCX, XLSX and PDF with a text layer (a scan with no text layer is skipped).

Privacy in the knowledge base

Notes and Files retain their own permissions: belonging to a Workspace does not grant access to them. The Assistant’s responses will only include documents that you can already open - your own documents and those shared with you. This also applies to administrators. A deleted document (bin / trash) is removed from the knowledge base during the next update.

Troubleshooting

“Error: invalid API key”

Check that the pasted key is correct. Make sure your account with the provider has active billing/limits.

“The model doesn’t support chat mode”

The selected model is not a chat model. Return to the AI settings and select a different model from the list.

“Rate limit exceeded”

Your account with the provider has reached its request limit. Wait a moment or upgrade to a higher plan.

The model list is empty

Click the refresh button (↻) next to the model list. If it's still empty - check the API key and your Internet connection.