Clanker
autonomous systems engineering cli agent for any cloud environment: AWS, GCP, Cloudflare, etc
Last verified:
What is Clanker?
Clanker Cloud is a local-first AI Ops Platform and desktop app for infrastructure operations. It lets teams ask questions about live cloud and Kubernetes environments in plain English, inspect topology and cost signals, review change plans, and approve execution from one private workspace. The app runs locally on macOS, Windows, and Linux, using your existing cloud credentials and AI keys without sending them to a hosted SaaS layer.
Key features include Deep Research for multi-provider scans that identify cost waste, security misconfigurations, bottlenecks, and resilience gaps; Kubernetes troubleshooting across EKS, GKE, and AKS; plain-English querying of live infrastructure; reviewed change plans before apply; and maker mode for explicit execution approval. It supports AWS, GCP, Azure, Tencent Cloud, Kubernetes, Cloudflare, Hetzner, DigitalOcean, Vercel, Supabase, Railway, Fly.io, Verda, GitHub, Sentry, and Datadog. The platform is built on the open-source Clanker CLI and supports bring-your-own AI keys from OpenAI, Anthropic, Cohere, Gemini, Mistral AI, Hugging Face, Perplexity, Ollama, llama.cpp, and Codex, Claude Code, GitHub Copilot, OpenClaw, Hermes agent systems.
Clanker Cloud is designed for builders shipping AI-built apps to production, lean DevOps and platform teams running multi-cloud infrastructure, AI researchers deploying and training their own systems, and founders or full-stack engineers who need production infrastructure without deep cloud specialization. It is especially suited for teams that want local credential custody, provider-agnostic workflows, and explicit approval before any infrastructure changes.
Clanker pricing
Pricing model: Freemium
Clanker Cloud is free during the current beta until July 2026, including desktop downloads and the local-first infrastructure workspace. After beta, planned tiers include: Hobby (free for personal use, side projects, and individual exploration), Lite ($5/month for lightweight desktop use), Pro ($20/month for builders and operators using production workflows), Business ($200/month for teams with shared infrastructure work and extra support), and Enterprise (contact-us pricing via [email protected] for larger organizations and custom support). BYOK AI costs are billed directly by your chosen AI provider (Anthropic, OpenAI, Google, Cohere, or local models) and are not included in the subscription. Cloud provider usage, Kubernetes costs, and infrastructure bills remain with the operator's own accounts.
Clanker pros
- Local-first desktop app keeps credentials on your machine
- No hosted SaaS layer required for infrastructure workflows
- Bring-your-own AI keys with no token markup from Clanker
- Deep Research scans cost, security, resilience, and topology in one pass
- Parallel AI agents analyze infrastructure across multiple providers simultaneously
- Evidence-backed findings with severity, affected resources, and cost impact
- Read-only by default with explicit maker mode for execution
- Reviewed change plans show intended impact before apply
- Supports AWS, GCP, Azure, Tencent Cloud, Kubernetes, Cloudflare, Hetzner, DigitalOcean, Vercel, GitHub, and more
- Kubernetes troubleshooting in plain English for EKS, GKE, and AKS
- One-minute setup with existing cloud and cluster credentials
- Open-source CLI engine under the desktop app for inspection
- Works with Codex, Claude Code, GitHub Copilot, OpenClaw, Hermes agent systems
- Export full scan reports as JSON or Markdown to share with team
- Mobile agent console for managing infra on your phone
- Cost exploration across connected providers and model usage
- Custom agents can be built in Python or Node.js for infrastructure
Clanker cons
- Currently in beta until July 2026 with planned paid tiers after
- AI model costs billed separately through your own provider, not bundled
- Ansible and Slurm support planned but not yet available
- Desktop app only (macOS, Windows, Linux), no web-based interface
- Maker mode requires explicit operator approval for every change
- Requires bringing your own cloud credentials and AI keys
- Early alpha CLI with active development still ongoing
- Enterprise pricing requires contact with sales, not self-serve
Frequently asked questions about Clanker
What is Clanker Cloud?
Clanker Cloud is a local-first desktop app for shipping with AI, gathering live context, troubleshooting, and managing cloud and Kubernetes environments. It lets teams ask questions about live environments, inspect topology and cost signals, review change plans, and approve execution from one workspace.
Do my cloud credentials leave my machine?
No. Your credentials stay on your machine instead of going through a hosted third-party service. Clanker Cloud runs private, read-only infrastructure checks without handing privileged credentials to a hosted copilot layer.
Which providers are supported?
Clanker Cloud supports AWS, GCP, Azure, Tencent Cloud, Kubernetes, Cloudflare, Hetzner, DigitalOcean, Vercel, GitHub workflows, Supabase, Railway, Fly.io, Verda, Sentry, and Datadog. Ansible and Slurm are planned next rather than current support.
Who is this for?
There are three main paths: builders shipping AI-built apps to production, lean DevOps teams running multi-cloud infrastructure, and research teams deploying, training, and running their own systems. It is also a strong fit for founders and full-stack engineers who need production infrastructure without deep cloud specialization.
Can I bring my own models and keys?
Yes. Clanker Cloud supports bring-your-own-keys, so you can use your own AI providers from OpenAI, Anthropic, Cohere, Gemini, Mistral AI, Hugging Face, Perplexity, Ollama, and llama.cpp without token markup. Model usage is billed directly by your chosen provider.
How does approval work before infrastructure changes?
Agents generate a plan showing what will change, and operators approve execution before anything happens. Clanker Cloud checks live context first, generates plans second, and only makes changes in explicit maker mode when you are ready.
Does Clanker Cloud support Kubernetes?
Yes. Kubernetes support spans EKS, GKE, and AKS for plain-English troubleshooting and cluster inspection. You can connect an existing kubeconfig, scan deployed resources across namespaces, and ask questions about pod status, service topology, recent events, and resource utilization.
What is Deep Research?
Deep Research fans out across every connected provider, runs parallel analysis with multiple AI models and specialised subagents, and returns prioritised findings covering cost drivers, misconfigurations, bottlenecks, and resilience gaps. Each finding includes severity, affected resources, evidence sources, estimated cost impact, and a concrete action label.
How fast is the setup?
Install the desktop app and connect your credentials in about one minute on macOS, Windows, or Linux. Then move straight into repo analysis and deploy planning. No hosted SaaS layer or long setup flow is required.
Is Clanker Cloud free?
Yes. Clanker Cloud is free during the current beta until July 2026. After beta, Hobby is planned as the free personal tier for personal use and side projects. Paid tiers include Lite at $5/month, Pro at $20/month, and Business at $200/month.