Skip to main content

​
Jinba

Shanone’s Jinba integration gives your agent 1 tool: running a published Jinba Flow workflow with custom input arguments. Jinba Flow itself is a no-code, natural-language workflow builder — Shanone’s role is simply to trigger workflows you’ve already built and published there.

​
Getting Started

1

Get your Jinba API Key

Generate a key from Settings → API in your Jinba account.
2

Add it to Shanone

Open Integrations in the Shanone dashboard, select Jinba, and paste in your API Key.
3

Retry your request

Once saved, shanone_execute_tool calls for jinba_run_workflow will succeed.

​
Available Tools

Shanone provides 1 tool for Jinba:
jinba_run_workflow — execute a published Jinba Flow workflow by its UUID, optionally passing a JSON array of {name, value} input arguments defined by the workflow’s INPUT_ configuration

​
Common Use Cases

Chat-triggered automation

Let a chat request kick off a pre-built Jinba workflow instead of rebuilding logic elsewhere

Parameterized runs

Pass different input arguments to the same published workflow for different requests

Cross-tool orchestration

Use a Jinba workflow as one step in a larger agent task that spans multiple integrations

Long-running task delegation

Hand off complex multi-step logic to a workflow built visually in Jinba rather than in-prompt

​
Troubleshooting

The workflow must be published in Jinba Flow before it’s runnable via the API — a workflow only saved as a draft in the editor won’t be reachable, and double-check the UUID from the Jinba Flow dashboard.
args must be a JSON string representing an array of {"name": ..., "value": ...} objects — check the workflow’s INPUT_ tools configuration in Jinba to see which argument names it expects.
Execution time depends on workflow complexity; the underlying request allows up to 120 seconds, so a complex multi-step workflow may legitimately take that long before returning a result.