Production machine learning requires coordinated pipelines, infrastructure, governance, deployment, and monitoring. This skill provides structured MLOps architecture guidance across major clouds and toolchains.

Agent telemetry setup can be error-prone when plugins, API keys, and endpoints must align. This skill guides Manifest installation, configuration, verification, and troubleshooting.

Production ML systems require coordinated data, training, deployment, and monitoring plans. This skill structures that work into practical MLOps phases.

Complex product work can require coordination across planning, coding, testing, and operations. Loki Mode provides a structured multi-agent workflow with task queues, checkpoints, and review steps.

Linux incidents often start with broad symptoms and scattered evidence. This skill provides a phased workflow for collecting facts, analyzing resources, checking logs, testing networking, and validating fixes.

Linux administrators need repeatable scripts for backups, monitoring, and maintenance. This skill provides Bash templates and guidance for common operational tasks.

Linux teams need repeatable scripts for common administration work. This skill provides structured Bash templates for backups, monitoring, logs, users, networking, and automation.

LLM teams need visibility into traces, cost, latency, quality, and prompt changes. This skill guides Claude, Codex, and Claude Code users through practical Langfuse instrumentation and evaluation workflows.

Kubernetes platforms are difficult to design securely across teams, clusters, and cloud providers. This skill guides architecture, GitOps delivery, observability, resilience, and platform engineering decisions.

Kubernetes releases require many coordinated artifacts and checks. This skill guides container preparation, manifests, Helm charts, service mesh, security, observability, and deployment validation.

Kubernetes manifests are easy to misconfigure when teams write them from memory. This skill turns requirements into structured YAML guidance with security, probes, resources, and service patterns.

Generic incident procedures slow triage and create inconsistent decisions during outages. This skill creates structured runbooks with checks, mitigation steps, escalation paths, and communication templates.

Hybrid environments create difficult choices across platforms, networks, security, compliance, cost, and operations. This skill structures those choices into practical architecture plans and validation steps.

Large machine learning workloads need costly hardware and reliable persistence. This skill prepares Hugging Face Jobs with suitable compute, secrets, schedules, and Hub storage.

Hugging Face Hub operations span many commands, resources, and safety considerations. This skill selects suitable `hf` workflows and explains their effects.

Creating consistent Helm charts requires careful metadata, values, templates, dependencies, and validation. This skill provides reusable structures and practical checks for Kubernetes packaging.

Committing and publishing routine changes can require several Git commands. This skill stages all changes, creates one commit, and pushes the current branch.

Complex Git changes can reach review without consistent testing, commit structure, or deployment checks. This skill organizes review, validation, commits, branches, and pull request preparation.

GitLab operations require repeated navigation, filtering, and identifier lookup across projects. This skill guides Claude, Codex, and Claude Code through Rube MCP workflows for repositories, pipelines, issues, and users.

GitHub maintenance requires many separate searches, checks, and updates. This skill coordinates repository workflows through Rube MCP with practical safeguards and validation guidance.