Triage failed egg agent Pods and produce prioritized reports.
npx skills add https://github.com/jwbron/egg --skill agent-diagnose
Please help me install this Agent Skill. Skill: agent-diagnose Source: https://github.com/jwbron/egg/tree/main/skills/agent-diagnose Command: npx skills add https://github.com/jwbron/egg --skill agent-diagnose
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.
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.
Invoke the agent-diagnose skill with the pipeline_id and container_id to generate a prioritized triage report.
High-intent search queries and answers about installing and using this skill.
Diagnosing a failed Kubernetes agent pod involves collecting the Job spec, Pod status, redacted logs, recent Events, and an egress probe to produce a prioritized triage report. This skill surfaces root causes like missing credentials, gateway 403 errors, or NetworkPolicy blocks for agent pods in the egg-agents namespace.
A gateway 403 or container-id lookup 404 error indicates an authentication or routing failure within your agent deployment. This skill applies a pattern-based classifier to redacted container logs and deployment context to surface and prioritize these specific root causes in a triage report.
To check if a NetworkPolicy is blocking egress traffic, this skill runs a network isolation probe against the failed agent pod using deployment context tools. It validates egress connectivity and surfaces NetworkPolicy blocks as a prioritized root cause in the triage report.
Yes, you can safely triage agent failures using container logs and environment variables because this skill applies a redaction helper to assemble redacted evidence. It collects environment data and logs via MCP primitives while ensuring sensitive information is masked before generating the triage report.
Diagnosing a single failed agent pod targets the specific root causes of one pod, whereas a full deployment diagnosis evaluates the broader pipeline health. This skill serves as the per-agent counterpart to deployment-diagnose, focusing on individual agent pods across roles like coder, tester, documenter, and reviewer.
Troubleshooting agent pod failures across roles like coder, tester, documenter, or reviewer requires collecting pod status, redacted logs, and recent events. This skill applies a pattern-based classifier to this evidence to produce a prioritized triage report for any failed agent role in the egg-agents namespace.
popoffvgTriage Platforma Kubernetes deployments with port-forwarding, endpoint checks, and pod diagnostics.
AvaTar-ArTsDiagnose and fix ToolUniverse tools with targeted tests and verification.
xg-gh-25Execute deterministic health checks on SwarmAI subsystems and summarize pass/fail results for CI workflows.
BjzyLabsQuery and analyze Grafana Loki logs using LogQL via logcli.
