mirror of
https://github.com/n8n-io/n8n.git
synced 2026-11-05 15:57:59 +08:00
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
38 lines
1.7 KiB
JSON
38 lines
1.7 KiB
JSON
{
|
|
"conversation": [
|
|
{
|
|
"role": "user",
|
|
"text": "I'd like to pull workflow metadata from another n8n instance into a data table on this one. Mostly for reporting."
|
|
},
|
|
{
|
|
"role": "assistant",
|
|
"text": "Which instance, and what columns should the data table have?"
|
|
},
|
|
{
|
|
"role": "user",
|
|
"text": "Source instance is https://wonderman.users.n8n.cloud/. The data table is called 'workflows' and should have instanceId, workflowId, workflowName, tags."
|
|
},
|
|
{
|
|
"role": "assistant",
|
|
"text": "Got it. Should it re-fetch on a schedule? And how should it handle workflows that already exist in the table — re-insert, or update in place?"
|
|
},
|
|
{
|
|
"role": "user",
|
|
"text": "Yes, run on a schedule. And it must update in place — no duplicate rows when the same workflow comes through twice. Use upsert."
|
|
}
|
|
],
|
|
"messageBudget": 6,
|
|
"complexity": "complex",
|
|
"tags": ["build", "schedule", "data-table", "n8n-api"],
|
|
"triggerType": "schedule",
|
|
"datasets": ["pr", "full"],
|
|
"executionScenarios": [
|
|
{
|
|
"name": "happy-path",
|
|
"description": "Data table is populated",
|
|
"dataSetup": "Workflow data:\n- instanceId: 123, workflowId: abc, workflowName: 'Bob workflow', tags: bob, lovely\n- instanceId: 123, workflowId: def, workflowName: 'Raving workflow', tags: dance, boogie\n- instanceId: 456, workflowId: xyz, workflowName: 'Sir Bob', tags: knight, bob\n- instanceId: 789, workflowId: pqr, workflowName: 'Algebra', tags: maths, blackboard\n",
|
|
"successCriteria": "The workflow executes without errors. A data table called 'workflows' is created and contains the specified data. The workflow contains a mechanism to avoid inserting duplicate workflows (e.g. by using upsert rather than insert)."
|
|
}
|
|
]
|
|
}
|