Detect LLM prompt regressions with Data Tables, OpenAI APIs, and Telegram
Ready for 3.0Ready for 3.0. This template uses no nodes that n8n 3.0 removes. No removed nodes, versions or helpers found. Static check, last run 2026-10-19.
What this workflow does
This workflow has 30 nodes and runs on a schedule (every day at 08:00). Along the way it connects Telegram.
How it runs
Trigger: On a schedule (every day at 08:00).
- Config (Edit Fields (Set)): sets 8 fields: apiBaseUrl, judgeModel, maxCases…
- Load profiles (Data table): get
- Load test cases (Data table): get
- Any data? (If): checks a condition and branches into true / false
- Build demo profiles (Code): runs custom JavaScript code
- Resolve profile (Code): runs custom JavaScript code
- Seed profiles (Data table): uses Data table
- Case valid? (If): checks a condition and branches into true / false
- Build demo cases (Code): runs custom JavaScript code
- Call model API (HTTP Request): calls an HTTP API (POST)
…plus 18 more steps. See the full node list below.
Services it uses
Before you upgrade to n8n 3.0
Nothing in this workflow is removed in n8n 3.0, so it should import and run after the upgrade.
Nodes used (9)
- Schedule Trigger
n8n-nodes-base.scheduleTrigger - Edit Fields (Set)
n8n-nodes-base.set - Data table
n8n-nodes-base.dataTable - If
n8n-nodes-base.if - Code
n8n-nodes-base.code - No Operation, do nothing
n8n-nodes-base.noOp - HTTP Request
n8n-nodes-base.httpRequest - Merge
n8n-nodes-base.merge - Telegram
n8n-nodes-base.telegram
From the author
Quick overview Continuous quality checks for your production LLM prompts. Keeps a golden test set in n8n Data Tables, runs it on a schedule against any OpenAI-compatible API, compares each run to a blessed baseline, and pings Telegram only when quality…