Agentic Ops Harness | Superpowers Like AI Harness for Ops. Currently works with Hermes Agent
This repository also configures its own agents. See what aoh tells them →
Skill Claude CodeCodex
Use when creating or revising reusable Agentic Ops Harness packs, ops skills, workflows, agent roles, model profiles, runtime requirements, evals, or runtime adapter mappings.
Skill Claude CodeCodex
Use when you've drafted or improved a skill locally and want to publish it to a shared pack — walks through validating the draft, finding the target pack, running the promote script, and reviewing the staged diff before it ships.
Skill Claude CodeCodex
Use when diagnosing Docker disk usage, dangling images, stopped containers, volumes, builder cache, or safe cleanup options on a local machine.
Skill Claude CodeCodex
Use when asked how healthy a cluster or namespace is — summarizes node readiness, deployment availability, pod restarts, warning events, and endpoints into a concise health report.
Skill Claude CodeCodex
Use when nodes are NotReady or the whole cluster looks unhealthy — inspects node conditions, CNI/kube-system pods, and capacity to find what is keeping the kubelet or network from being ready.
Skill Claude CodeCodex
Use when pods are stuck Pending — inspects scheduling events, resource requests vs node allocatable, taints/tolerations, and PVC binding to find why the scheduler cannot place them.
Skill Claude CodeCodex
Use when pods are in CrashLoopBackOff, ImagePullBackOff, ErrImagePull, or OOMKilled — collects describe/logs/events and identifies the failure class before recommending a fix.
Skill Claude CodeCodex
Use when planning, validating, or automating deployment workflows, release checks, rollback readiness, and post-deploy verification.
Skill Claude CodeCodex
Use when planning and validating safe deployment automation — orchestrates deployment planning, terraform plan review, and post-release health verification.
Skill Claude CodeCodex
Use when triaging failed, slow, or expensive ML training jobs, GPU utilization, data/input errors, checkpoints, and retry strategy.
Skill Claude CodeCodex
Use when triaging failed or expensive model training jobs — orchestrates training job triage, then platform health checks, then a safe retry strategy.
Skill Claude CodeCodex
Use when triaging platform reliability issues — orchestrates service health reporting, then Docker disk diagnostics, then the smallest safe next action.
Skill Claude CodeCodex
Use when collecting read-only service health signals, recent failures, dependency status, and concise operational risk summaries.
Skill Claude CodeCodex
Use when reviewing Terraform plans, infrastructure diffs, blast radius, drift, or safe IaC automation steps.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: