Which n8n nodes do AI agents use most, and which trigger nodes start their workflows?
The most used node in AI-built n8n workflows is Code, present in 77% of last week's workflows, followed by HTTP Request (59%) and Webhook (49%). The median workflow has 7 nodes.
- Last updated
- Oct 23, 2026
- Data window
- Oct 5–11, 2026
- Author
- Romuald Członkowski
- Code77%
- HTTP Request59%
- Webhook49%
- IF38%
- Schedule Trigger25%
- Manual Trigger19%
- Edit Fields (Set)18%
- Respond to Webhook16%
- Execute Workflow15%
- Execute Workflow Trigger14%
- Postgres13%
- Data Table12%
- Google Sheets9%
- Telegram7%
- Switch7%
The top three are general-purpose tools: code, API calls and webhook entry. Application integrations come lower, led by Google Sheets, Postgres and Telegram. 15% of workflows call another workflow, so composition from sub-workflows is the norm.
Trigger nodes: Webhook 43%, schedule 21%, app event 19%, manual 14%, form 2%. Webhooks start half of AI-built workflows, which means n8n mostly serves as the integration layer between systems.
The corpus holds 1,190,875 workflows built or modified by AI agents since March 2026; the shares above cover last week's workflows, and the history below tracks how the top of the ranking shifts from issue to issue.
History
One row per weekly issue. The table grows with every update.
| week | #1 | #2 | #3 | #4 | #5 |
|---|---|---|---|---|---|
| 2026-W41 | Code 77% | HTTP Request 59% | Webhook 49% | IF 38% | Schedule Trigger 25% |
| 2026-W40 | Code 75% | HTTP Request 61% | Webhook 50% | IF 39% | Schedule Trigger 25% |
| 27 Sep 2026: telemetry source changed | |||||
| 2026-W37 | Code 73% | HTTP Request 58% | Webhook 50% | IF 36% | Schedule Trigger 23% |
| 2026-W36 | Code 73% | HTTP Request 53% | Webhook 47% | IF 32% | Schedule Trigger 24% |
| 2026-W35 | Code 74% | HTTP Request 60% | Webhook 55% | IF 34% | Schedule Trigger 24% |
On 27 September 2026 I moved n8n-mcp telemetry to new infrastructure. Versions older than 2.90.0 stopped reporting that day, so figures from before and after it come from different sets of installs. I don't compare one with the other.
Methodology · Data and charts may be reused with attribution (CC BY 4.0).