How to Fix n8n AI Agent Tool Calls Not Saved in the Memory Node
n8n's AI Agent never writes the tool_calls array to Memory, so the model imitates a fake history and stops calling tools. Here are the two fixes that work today.
Topic
How To
Tactical how-to / step-by-step / troubleshooting
Fix LinkedIn "Requested Version Is Not Active" in Zapier and Make
LinkedIn retires its API version every year, and Zapier's and Make's LinkedIn connectors break when it does. Here's the same-day Zapier fix, and why Make has none yet.
Zapier and monday.com Webhooks Return 500 Error After monday's Federated API Migration: How to Fix It
Why a working Zapier-to-monday.com webhook suddenly starts throwing 500 errors, the header that fixes it, and why that fix isn't permanent.
Claude Code AskUserQuestion "No Response After 60s": How to Fix the AFK Timeout
Why Claude Code's AskUserQuestion silently answers itself after 60 seconds, and the version and config fix - plus the terminal-focus gotcha most write-ups skip.
How to fix Power Automate desktop flow NoCandidateMachine error
Decode the NoCandidateMachine sub-error codes and fix the session or machine state blocking your desktop flow, with the 12-hour queue timeout explained.
How to set up Entra Service Principal auth for n8n Microsoft nodes
Run n8n's Microsoft nodes unattended: the full Entra app registration and Service Principal credential setup for n8n 2.30, plus the node behaviour change to plan for.
Fix UiPath Studio 2026.0.196 "workflow missing" XAML errors
Studio 2026.0.196 flags present XAML files as missing after the update. Here's the cache-clear fix and the patched build that ends it.
Add MCP Tools to a Make AI Agent (Add MCP Button)
Make AI Agents can finally call external MCP server tools next to native modules - here is where the Add MCP button lives and how the connection works.
How to fix Make's Retry error handler (formerly Break) not triggering
Make's Retry error handler (formerly Break) silently skips HTTP module failures until you flip one toggle on the HTTP module itself.
How to Deploy an AI Agent to Production on AWS Lambda (Serverless)
Lambda isn't a cheaper container: no warm process for state, a 900-second ceiling, and streaming that's native only on Node.js. Here's how to ship an agent around all three.