agent-diagnose

Community

Triages a failed egg agent Pod for root causes.

Authorjwbron
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Triage a single failed egg agent Pod to identify root causes quickly by collecting essential evidence (Job spec, Pod status, redacted logs, recent Events, redacted environment, and an egress probe) and applying a pattern-based classifier to prioritize findings.

Core Features & Use Cases

This skill composes MCP primitives (list_containers, get_container_logs) with deployment tools (get_deployment_context, validate_network_isolation) and a redaction helper (orchestrator/redaction.py) to produce a prioritized triage report for a specific failed agent. It is used when an agent Pod in egg-agents exits non-zero across roles (coder, tester, documenter, reviewer) and the pipeline is stuck or failed, or when you need to understand why the agent failed (missing credentials, gateway 403, container-id lookup 404, sandbox stack traces, or a NetworkPolicy block). It is the per-agent counterpart to /deployment-diagnose.

Quick Start

Invoke the agent-diagnose skill with the pipeline_id and container_id to generate a prioritized triage report.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: agent-diagnose
Download link: https://github.com/jwbron/egg/archive/main.zip#agent-diagnose

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.