Mistral AI integration

What it does

The Mistral AI integration lets your agency invoke Mistral chat completions, create embedding vectors with mistral-embed, generate fill-in-the-middle code completions with Codestral, and inspect the model catalog from any workflow. Connect a client's Mistral API key once and your workflows can summarize incoming form submissions, classify support tickets, draft European-data-residency-friendly responses, embed knowledge bases for retrieval, or autocomplete code snippets, all without leaving TaskJuice.

Connect a Mistral AI account

1. Open your workspace in TaskJuice and navigate to Connections.
2. Choose Mistral AI and click Connect.
3. In a new tab, open the Mistral console API keys page signed in as the client (or as your agency, if the client has delegated key creation).
4. Click Create new key, give it a name that identifies the workspace, and copy the value.
5. Paste the key into TaskJuice and save the connection.

To rotate or revoke the key later, return to the Mistral console, delete the entry, create a new key, and update the TaskJuice connection.

Triggers

Mistral AI does not publish a public webhook surface, so the integration is action-only. Use a Schedule or another app's trigger to invoke Mistral actions inside a workflow.

Actions

• mistral/create-chat-completion generates a chat completion against a chosen model (for example, mistral-small-latest, mistral-large-latest) with a list of messages and optional temperature, max tokens, top-p, and response-format controls.
• mistral/create-embedding returns an embedding vector for an input string using the mistral-embed model.
• mistral/create-fim-completion generates a fill-in-the-middle code completion using a Codestral model, taking a prompt prefix and optional suffix.
• mistral/list-models returns the model catalog available to the authenticated key, useful for surfacing a model picker to end users or branching on model availability.

Known limitations

• The integration calls the Mistral REST API with the connected key. Per-key rate limits, monthly spend caps, and model-access entitlements are governed by the client's Mistral account, not by TaskJuice. When Mistral returns a 429, TaskJuice surfaces it as a retryable rate-limit error and honors the recommended backoff.
• Streaming responses are not exposed. Each chat or FIM completion runs as a single request and resolves only after the full response is returned by Mistral.
• Embedding requests accept a single input string per call. For batch embedding, fan out across a Loop node so each item runs as its own request.
• The integration is action-only. To react to events in real time, use a Schedule trigger, a polling trigger from another app, or an inbound webhook from a different source and call Mistral inside the workflow.