Skip to main content

​
Gumloop

Shanone’s Gumloop integration gives your agent 6 tools for triggering and managing Gumloop flows (no-code AI automations) — from discovering what’s available to running a flow and tracking it to completion.

​
Getting Started

1

Get your Gumloop API Key

Generate a key from Settings in your Gumloop account.
2

Add it to Shanone

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

Retry your request

Once saved, shanone_execute_tool calls for gumloop_* tools will succeed.

​
Available Tools

Shanone provides 6 tools for Gumloop, organized into these categories:
gumloop_list_flows, gumloop_get_flow_schema — list saved flows and inspect a flow’s required/optional inputs before running it
gumloop_run_flow, gumloop_kill_run — start a flow with input values, or stop one that’s taking too long
gumloop_get_run_status, gumloop_list_run_history — poll a specific run for completion, or review past runs for a flow

​
Common Use Cases

Automation triggering

Kick off a Gumloop flow (e.g. lead enrichment, content generation) from a chat request

Run polling

Start a long-running flow and poll until it reports DONE before using its outputs

Flow input discovery

Inspect a flow’s schema before running it to avoid missing required inputs

History review

Pull recent runs for a flow to debug a failure or check throughput

​
Troubleshooting

Call gumloop_get_flow_schema first with the flow’s saved_item_id to see exactly which input keys are required, then pass them in the inputs dict.
Some flows (especially ones involving web scraping or LLM calls) can legitimately take minutes; if it seems stuck, use gumloop_kill_run and re-run rather than waiting indefinitely.
Only saved/published flows are returned — a flow still open only in the Gumloop canvas editor without being saved as an item won’t appear until it’s saved.