> ## Documentation Index
> Fetch the complete documentation index at: https://darwin.so/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Gumloop

> Add Darwin capability discovery and durable work to a Gumloop flow.

Use a protected HTTP or code node to call Darwin. Keep Search as a read-only discovery step and represent each Action mutation as a separate, reviewable step.

Prefer Gumloop's MCP connection UI and follow the [public reusable-node recipe](https://github.com/darwin-studios/darwin-integrations/tree/main/templates/gumloop) so tools are discovered dynamically.

Preserve canonical result order and the selected capability revision. Store `actionId` so the flow can resume after waiting for a person, a provider connection, or a payment interaction.

Do not place API keys, provider tokens, or payment data in prompts, visible node labels, run logs, or generated output.

## Flow checkpoints

| Checkpoint           | Persist                                 |
| -------------------- | --------------------------------------- |
| After Search         | Canonical results and selected revision |
| After Start          | `actionId` and Start `requestId`        |
| Before a human wait  | Latest state and valid next actions     |
| After an interaction | Refreshed Action response               |

Do not hold one flow open indefinitely. Resume from durable state and call Get action before choosing the next node.