agent-investigate

Collect and correlate Docker logs, OTEL events, and lifecycle data for Claude agent executions.

Updated Sep 22, 2025

Search Tags:#debugging #troubleshooting #lifecycle #analyze #agent-execution #otel #docker-logs

Ready to run

npx skills add https://github.com/tinkermonkey/switchyard --skill agent-investigate

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-investigate
Source: https://github.com/tinkermonkey/switchyard/tree/main/.claude/skills/agent-investigate
Command: npx skills add https://github.com/tinkermonkey/switchyard --skill agent-investigate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Analyzes Claude agent executions to diagnose issues by collecting and correlating Docker logs, OTEL events, and lifecycle data.

Core Features & Use Cases

Retrieve and correlate Docker logs, OpenTelemetry events, and agent lifecycle timings for a given execution.
Inspect a specific execution by container name, task_id, or pipeline_run_id + agent_name.
Build a concise execution narrative including root cause analysis and recommendations.

Quick Start

Provide an execution identifier (container name, task_id, or pipeline_run_id + agent_name) and the skill will fetch logs, events, and lifecycle data for analysis.

Frequently Asked Questions about agent-investigate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I debug Claude agent executions using Docker logs and OpenTelemetry events?▼

To debug Claude agent executions, you analyze Docker logs and OpenTelemetry (OTEL) events by targeting a specific container name, task_id, or pipeline_run_id to collect and correlate lifecycle data for issue diagnosis.

What is the best way to correlate agent lifecycle timing with OTEL execution events?▼

Correlating agent lifecycle timing with OTEL execution events involves retrieving local endpoint data using a specific execution identifier like task_id, allowing you to build a concise narrative for root cause analysis.

Can I diagnose a specific agent execution using a pipeline_run_id and agent_name?▼

Yes, you can diagnose a specific agent execution by providing a pipeline_run_id combined with an agent_name, which fetches the corresponding Docker logs, OTEL events, and lifecycle timing for analysis.

How do I check the recent health of my Claude agent activity?▼

Checking recent Claude agent health involves analyzing recent activity logs and events without a specific execution identifier, applying lifecycle data collection to perform a quick diagnostic of overall agent performance.

Does agent troubleshooting require a specific container name to retrieve logs?▼

Agent troubleshooting does not strictly require a container name; you can retrieve logs and analyze executions by supplying either a task_id or a pipeline_run_id paired with an agent_name as your identifier.