Inspect ADK agent sessions, event flows, and LLM interactions.
npx skills add https://github.com/rahulgupta2018/agent-skills --skill adk-debug-rahulgupta2018
Please help me install this Agent Skill. Skill: adk-debug Source: https://github.com/rahulgupta2018/agent-skills/tree/main/skills/adk-debug Command: npx skills add https://github.com/rahulgupta2018/agent-skills --skill adk-debug-rahulgupta2018
This skill resolves complex issues in ADK agent development by providing a structured framework for inspecting event flows, debugging LLM interactions, and testing agent logic in both CLI and web environments.
Use the adk-debug skill to run the agent in verbose mode and output the event stream to a log file for inspection.
High-intent search queries and answers about installing and using this skill.
To debug an ADK agent, you can inspect session events and trace LLM request/response spans to pinpoint logic errors. The skill provides standardized logging to help identify bottlenecks or failures in tool execution and configuration.
ADK agent tool execution failures often stem from configuration mismatches or code errors within the agent logic. You can trace the specific LLM request and event flow using diagnostic endpoints to perform root cause analysis on the failure.
Yes, you can inspect ADK agent sessions in both CLI and web server environments. The skill supports interactive CLI testing via adk run and visual session inspection via the adk web REST API for multi-turn agent scenarios.
The best way to test multi-turn ADK agent scenarios is by using the interactive CLI testing or REST API-based web server monitoring. These diagnostic modes allow you to monitor event flow and analyze interactions across multiple turns.
To capture verbose event streams for ADK agent troubleshooting, run the agent in verbose mode and output the event stream to a log file. This facilitates structured root cause analysis of agent failures and tool execution errors.
kevinnguyen271090Build and deploy AI agents with Google ADK Python.
BBufAnalyze LLM torch-profiler traces to produce kernel, overlap, and fuse tables.
MinhHoangDonoBuild AI agent applications with Google ADK Python using tools, multi-agent composition, and workflows.
ddnettersQuery Langfuse REST API for LLM traces, generations, costs, and metrics.
