Developer Platform: MCP server availableDeveloper Platform: MCP server available, App SDK from Q4 2026

[ AI ]

Letta integration for your workflows.

Connect Letta to Collabase once and use 4 actions and 2 triggers in your workflows. No code needed, hosted in Switzerland.

Sign-inCustom credentials
CategoryAI
UpdatedSeptember 2026
LicenceOpen source (MIT)
Back to the directory

Letta in Collabase

Connect Letta to Collabase once and use 4 actions and 2 triggers in your workflows. No code needed, hosted in Switzerland.

[ About the service ]

Letta is the platform for building stateful agents: open AI with advanced memory that can learn and self-improve over time.

How to connect it

  1. Open the integrations in the workspace settings.
  2. Choose Letta and connect your account.
  3. Add a trigger or an action to the workflow.
  • For Letta Cloud: Get your API key from Letta Cloud
  • For Self-hosted: Leave API key empty and provide your server URL (e.g., http://localhost:8283)

[ AI ]

Related integrations

All integrations

Tools that often share a workflow.

Questions about Letta

What can I automate with Letta in Collabase?

Letta brings 4 actions and 2 triggers, for example Create Agent From Template, Create Identity, Send Message to Agent and Get Identities. You combine them in a workflow without code and mix them with every other integration.

How do I connect Letta to Collabase?

You open the integrations in the workspace settings, choose Letta and connect your account once, using custom credentials. After that all triggers and actions are available in the workflow.

Can Letta start a workflow?

Yes. Letta provides 2 triggers, for example New Agent and New Message. The workflow runs as soon as the event happens.

Does the Letta integration cost extra?

No. All integrations are included in every licence, in the Isolated Cloud and in Self-Hosted operation. There is no module price.

Test it against your own content. A proof of concept is ready within days.

We set up an instance with your content and you test it against real workflows. In a first conversation we clarify scope, operating model and timing.