Find CPU and memory hotspots in deployed Workers.
Identifies CPU and memory hotspots in deployed Workers and Durable Objects and recommends small, evidence-backed code changes.
When to use it
Use to investigate CPU usage, latency, allocation pressure, or memory growth in deployed workloads.
Give it a deployed workload to investigate; it returns profiling evidence, a minimal recommended change, and a way to check the effect.
Requires the Cloudflare account containing the deployed workload; the agent confirms the account with the user before profiling.
Requires a deployed Worker to profile; profiling a Durable Object also requires identifying its owning Worker, namespace, and instance.
This skill
Cloudflare Workers
Captures deployed workload profiles
Cloudflare Workers production profiling guide
Reads the production profiling guide