Inspect runtime variable values in production services using Datadog Live Debugger log probes.
npx skills add https://github.com/lauhon/pi --skill dd-debugger-lauhon
Please help me install this Agent Skill. Skill: dd-debugger Source: https://github.com/lauhon/pi/tree/main/skills/dd-debugger Command: npx skills add https://github.com/lauhon/pi --skill dd-debugger-lauhon
--fields output or jq-based extraction of snapshots, arguments, and locals. - Service Context Discovery: Verify environments, languages, and active instances before instrumenting to avoid targeting the wrong deployment. - Use Case: A user reports that an order endpoint receives malformed data. Place a probe on OrderService:processOrder with --capture "order.items[0].price", stream events for 60 seconds, inspect the live values, then delete the probe. ## Quick Start Ask the agent to place a Datadog live debugger probe on a specific method of your service and show the runtime argument values it captures.High-intent search queries and answers about installing and using this skill.
Use Datadog Live Debugger log probes placed on a target method with the pup CLI. Create a probe with capture expressions like --capture "user.name", then stream events with pup debugger probes watch to see live values.
Run pup debugger probes create with --service, --env, --probe-location in TYPE:METHOD form, and one or more --capture expressions. Add --ttl, --condition, or --template to control lifetime, filtering, and log message format.
The debugger supports Java, Python, and .NET services, specified via the --language flag as java, python, or dotnet. Language is auto-detected from symdb when not provided.
Check that the --from flag is not excluding recent events, since the default is now and only shows new events. Also verify the environment with pup debugger context and allow time for instrumentation, or use --wait to retry probe lookup.
Capture expressions like --capture "order.id" return only the specific values you request and are lighter-weight. A bare --capture flag takes a full snapshot of all arguments, locals, and return values, which is heavier and recommended only when you do not know which fields to inspect.
datadog-labsSearch Datadog Symbol Database to find probe-able methods for Live Debugger instrumentation.
sameer-thakkarCorrelate logs, traces, metrics, and code to diagnose production incidents.
elasticAssess APM service health by querying Observability APIs and Elasticsearch.
KumarAdi8Diagnose iOS crashes, memory leaks, and hangs using LLDB and Instruments.