By mckinsey
Automate the full development lifecycle of the Ark Kubernetes operator for AI agents: run Playwright dashboard tests, manage CVEs and OWASP vulnerabilities, generate CRDs, orchestrate OpenSpec feature workflows, provision isolated Kubernetes clusters, and sync documentation.
You are an expert Kubernetes CRD developer working on the ARK project. Your task is to generate a new Custom Resource Definition following ARK's established patterns and design guidelines.
Implement tasks from an OpenSpec change (Experimental)
Archive a completed change in the experimental workflow
Archive multiple completed changes at once
Continue working on a change - create the next artifact (Experimental)
Use this agent when designing architecture for new Ark features, planning component extensions, or evaluating technical approaches that need to align with existing codebase patterns. Examples: - User: "I need to add a new CRD for managing prompt templates" Assistant: "I'll use the ark-architect agent to design how this fits with existing Ark patterns." <launches ark-architect agent> - User: "How should we implement multi-tenant support in Ark?" Assistant: "This requires architectural planning. Let me use the ark-architect agent to analyze the codebase and design a solution." <launches ark-architect agent> - User: "We need to add webhook support to the executor service" Assistant: "I'll engage the ark-architect agent to examine existing webhook patterns in the operator and design a consistent approach." <launches ark-architect agent>
Orchestrates multi-phase Ark feature development using the task folder protocol. Manages task lifecycle, coordinates specialist agents (ark-architect, ark-prototyper), and tracks progress through objectives → architecture → prototype → outcome phases.
Use this agent to build minimal, verifiable prototypes of Ark features. It makes the smallest changes needed to demonstrate the architecture and user experience - minimal CRD updates, simple API endpoints, basic CLI commands, and dashboard UI. Examples: - User: "Prototype the ark-broker feature" Assistant: "I'll use the ark-prototyper agent to build a minimal working version." <launches ark-prototyper agent> - User: "Create a simple version we can test" Assistant: "Let me use the ark-prototyper to implement the minimum viable changes." <launches ark-prototyper agent>
Fix security vulnerabilities in Ark by researching CVEs, analyzing impact, proposing mitigations, implementing patches, and creating PRs. Use when the user reports CVE numbers or security issues that need fixing in Ark. Examples: - User: "Fix CVE-2025-55183 in Ark" Assistant: "I'll use the ark-security-patcher agent to research this vulnerability and apply a fix." <launches ark-security-patcher agent> - User: "The current version of golang has a vulnerability, fix it in ark" Assistant: "Let me use the ark-security-patcher agent to identify and fix this vulnerability." <launches ark-security-patcher agent> - User: "Patch the security issue in our dependencies" Assistant: "I'll engage the ark-security-patcher agent to analyze and remediate security vulnerabilities." <launches ark-security-patcher agent>
Research technical solutions by searching the web, examining GitHub repos, and gathering evidence. Use when exploring implementation options, evaluating technologies, or investigating how to solve a technical problem.
Reference documentation for the Agent2Agent (A2A) protocol. Use when building A2A servers or clients, configuring Ark A2AServer resources, debugging A2A communication, or answering questions about the A2A specification, Agent Cards, task lifecycle, streaming, extensions, or protocol bindings.
Analyze the Ark codebase by cloning the repository to a temporary location. Use this skill when the user asks questions about how Ark works, wants to understand Ark's implementation, or needs to examine Ark source code.
Design architecture for Ark features following existing patterns and principles. Use when planning new features, extending components, or evaluating technical approaches.
Guidance for developing the Ark Kubernetes operator. Use when modifying Go types, CRDs, controllers, or webhooks. Helps with CRD generation and Helm chart sync issues.
Consolidate open Dependabot PRs into a single integration branch. Use when the user asks to "consolidate dependabot", "merge dependabot PRs", "batch dependency updates", or mentions dependabot PR management.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
⚒️ arkA declarative, Kubernetes-native framework for building portable, scalable, and provider-agnostic agentic applications.
Features • Quickstart • Documentation
Ark is a declarative toolkit for building and hosting distributed AI agents. By defining what agents should do rather than how they do it, Ark eliminates vendor lock-in and ensures your applications stay adaptable as AI evolves.
Built on Kubernetes, Ark lets you deploy a dev-friendly cluster in minutes or scale agentic workloads across existing infrastructure. Leverage proven patterns for security, monitoring, and RBAC—avoiding bespoke overhead while maintaining a portable, production-ready foundation for your AI projects.
Ark is designed for rapid, democratic development of agentic systems. The entire stack is built on open-source Kubernetes technology designed for running distributed systems. It can run comfortably on a single developer's machine or be deployed into a Kubernetes cluster across multi-cloud and on-prem environments. Developers and operations teams have full visibility into the entire stack, from the highest to the lowest levels.
Because each workload is a declarative specification of agent behavior rather than proprietary code, teams can re-platform individual use cases onto specialized or proprietary stacks when needed, typically with minimal migration overhead.
You will need a Kubernetes cluster to install Ark into. You can use Minikube, Kind, Docker Desktop or similar to run a local cluster.
Ensure you have Node.js and Helm installed. Then run the following commands to install Ark:
# Install the 'ark' CLI:
npm install -g @agents-at-scale/ark
# Install Ark:
ark install
# Optionally configure a 'default' model to use for agents:
ark models create default
# Run the dashboard:
ark dashboard
In most cases the default installation options will be sufficient. This will install the Ark dependencies, the controller, the APIs and the dashboard. You can optionally setup a default model that will be the default used by agents. The install command will warn if any required dependencies are missing.
npx claudepluginhub p/mckinsey-ark-claudeBuild Vizro dashboards from concept to deployment. Enforces a 2-phase workflow covering requirements, layout design, visualization selection, implementation with Python, and testing.
Claude Code skill pack for Cast AI (18 skills)
Kubernetes and Helm operations - deployments, charts, releases, diagnostics
AIDLC Operations phase automation — self-improving loops, autonomous deploys, continuous evaluation, incident response, and cost governance on AWS. Humans approve at checkpoints; agents execute diagnosis, proposal, and remediation between gates.
Kubernetes container orchestration toolkit for writing production-ready manifests, managing clusters, and implementing cloud-native best practices.
Kubernetes cluster operations, health diagnostics, and operator-specific agents
Build, deploy, and operate AI agents on AWS. Skills for scaffolding agents with Amazon Bedrock AgentCore (Strands, LangGraph), connecting tools via Gateway and MCP, multi-agent and A2A orchestration, memory, Cedar policies, evaluation, observability, debugging traces and logs, and production hardening (inbound auth, IAM, rate limiting, cold-start tuning).