Work public
code-first-n8n
Five proof-of-concept workflows showing 96 percent token savings with Code-Mode versus traditional n8n AI agents. Fully TypeScript with n8nac.
What it is
Five proof-of-concept workflows that benchmark Code-Mode (HTTP requests inside a code node) against the traditional n8n AI-agent pattern (one LLM call per tool). The benchmark output: 96 percent token savings on representative pipelines.
The point is the benchmark. The code-first practice grew from this PoC into the n8n-project-template and the n8n-autopilot projects.
Stack
TypeScript with n8nac. Five workflows, one benchmark, zero euro LLM cost on the test set.