1 min lesson
Investigate the symptom yourself with an observability MCP
Use "At Cursor the observability stack is Datadog, and the speaker leans on the Datadog MCP heavily" to describe the practical rule.
Step 1 of 2
Investigate the symptom yourself with an observability MCPyou don't have to wait for an engineer
A paper cut often shows up as a vague symptom: one user's data looks wrong, a state seems impossible. At Cursor the observability stack is Datadog, and the speaker leans on the Datadog MCPModel Context Protocol. A standard that lets an AI agent pull in context from outside the repo, like Jira tickets or internal docs. Press Enter for the full definition. heavily - Datadog ships a Cursor/VS Code extension that installs the MCP for you. With it plugged in you hand Cursor a user ID or a wrong-data report and tell it to investigate. It sifts the logs, lines up timestamps, resolves what it finds against the actual code and comes back with a thorough analysis - often more thorough than you'd manage by hand, and a far more ergonomic interface than clicking around the Datadog UI.
“I use the Datadog MCPModel Context Protocol. A standard that lets an AI agent pull in context from outside the repo, like Jira tickets or internal docs. Press Enter for the full definition. extremely heavily. They actually have a Cursor or a VS Code extension that you can install that should also install their MCP.”