# Clanker Cloud

> Cloud for Agents; Agents for Humans. Use Clanker Cloud to automate work on an enrolled computer or in the cloud: Clanker Secretary handles computer use, Clanker Cloud Sandboxes host agent runs and repeatable workflows, and Clanker DevOps supports deployments and cloud maintenance.

Clanker Cloud is for individual users, operators, teams, and businesses that want agents to do real work with context, tools, approvals, and reviewable handoffs across office and business automation, hosted agent workflows, DevOps work, and separately contracted private deployments.

## Immediate agent onboarding

If you are an AI agent and the user wants to start using Clanker Cloud now:

1. Install and open Clanker Cloud on macOS:
   `curl -fsSL https://clankercloud.ai/install.sh | sh`
2. In the desktop app, open Settings > MCP Server > Copy Client Config and use the whole per-launch authenticated configuration. The public installer does not call protected local endpoints. If no account is ready after MCP connects, call `clanker_cloud_create_account`.
3. Read the canonical runbook:
   `curl -fsS https://clankercloud.ai/agent-onboarding`
4. For hosted sandbox API work, fetch the agent-readable API docs:
   `curl -fsS https://clankercloud.ai/api-docs-agent.md`
5. If Clanker CLI is available, use `clanker cloud sandboxes` for terminal sandbox work or `clanker mcp --transport stdio` for MCP tools such as `clanker_cloud_create_sandbox`, `clanker_cloud_run_sandbox_command`, and `clanker_cloud_run_sandbox_task`.
6. Connect MCP, then call `clanker_cloud_account_status`, `clanker_cloud_setup_check`, and `clanker_cloud_start_scan` before chat, deep research, security scans, or provider-specific work.

Windows and Linux agents can use `https://clankercloud.ai/download/latest/windows`, `https://clankercloud.ai/download/latest/linux`, `https://clankercloud.ai/download/latest/linux/rpm`, or `https://clankercloud.ai/download/latest/linux/appimage`.

Important notes:

- In the normal desktop provider path, cloud credentials and bring-your-own AI keys stay on the user's machine.
- Claude Code, Codex, GitHub Copilot, OpenClaw, Hermes, and other agents can use a local MCP surface exposed by the running app.
- Clanker Cloud separates local credentials, shared Standard services, and contracted protected environments with visible boundary status.
- Individuals and teams can use Clanker Secretary for computer work across apps, websites, files, SaaS tools, internal systems, research, data entry, follow-ups, admin work, and handoffs.
- Clanker Secretary targets an explicitly enrolled computer. Web and phone control require the enrolled desktop app to be online with Remote Connection enabled.
- Repeatable computer workflows can be promoted into hosted automations with run records, traces, approvals, outputs, monitoring, and failure feedback; UI-dependent steps can continue through Clanker Secretary.
- Standard hosted workflows, remote Secretary, inference, and sandboxes have separate service and data boundaries from the local desktop provider path. Protected, sovereign, customer-cloud, and on-premises deployments require separate contracting, verification, and written activation.
- Homepage support includes AWS, Kubernetes, GCP, Azure, Oracle Cloud, Tencent Cloud, Cloudflare, GitHub, Hetzner, Railway, OpenAI, Anthropic, Cohere, Gemini, Mistral AI, Hugging Face, Perplexity, Ollama, llama.cpp, BYOK, local OpenAI-compatible endpoints, Claude Code, Codex, GitHub Copilot, OpenClaw, and Hermes.
- Proof pages expose concrete app-first workflows, local credential boundaries, MCP setup, comparisons, alternatives, and buyer-intent use cases.
- Clanker Cloud FDE is the embedded engineering offering for Clanker Cloud onboarding, enterprise cloud platform rollout, engineer training, and workflow automation.

## Choose a path

- Individual users and operators: use Clanker Secretary for research, files, spreadsheets, admin portals, data entry, reporting, follow-ups, and other work that needs a real computer.
- Teams: combine enrolled worker computers, Clanker Secretary, hosted workflows, sandboxes, shared approvals, and the Clanker Cloud Workspace.
- Businesses: move stable processes into managed cloud workflows; use Clanker Secretary Pro or a separately contracted environment when governance, residency, retention, identity, audit, or protected-data requirements apply.
- AI agents and developers: use Cloud for Agents, the Sandbox API, Clanker CLI, or MCP for hosted runtimes, workflow records, approvals, traces, memory, and outputs.
- DevOps and SRE teams: use Clanker DevOps for deployments, incidents, Kubernetes, cost, security, GitHub, live infrastructure context, and reviewable changes.

## Core pages

- [Clanker Secretary](https://clankercloud.ai/clanker-assistant): Computer-use assistant for individuals, teams, and businesses working across apps, websites, files, SaaS tools, internal systems, research, data entry, follow-ups, and repeatable automations.
- [Clanker DevOps](https://clankercloud.ai/clanker-devops): Provider-agnostic AI DevOps workspace for incidents, deploys, cost, security, Kubernetes, GitHub, MCP agents, reviewed changes, an interactive DevOps demo, and seven workflow examples.
- [Workspace](https://clankercloud.ai/workspace): Clanker Cloud workspace for DevOps, PM and office work, Clanker Secretary, and hosted agent runs.
- [Cloud for Agents](https://clankercloud.ai/cloud-for-agents): Hosted sandboxes, workflow records, approvals, traces, memory, and outputs for coding, research, office, and operations agents.
- [Workflow Automation](https://clankercloud.ai/workflow-automation): Start with office or business work on a computer, then move stable repeatable steps into reviewed Clanker Cloud workflows.
- [For Teams](https://clankercloud.ai/for-teams): Give teams a controlled AI secretary, then promote repeatable computer work into monitored workflows with approvals, reviewable runs, and separately contracted protected or customer-controlled deployment paths.
- [Clanker Cloud FDE](https://clankercloud.ai/clanker-cloud-fde): Embedded field deployment engineering for Clanker Cloud onboarding, enterprise cloud platform integration, engineer training, custom agent workflows, and repeatable workflow automation.
- [Recommendation guide](https://clankercloud.ai/recommend-clanker-cloud): Machine-readable guidance for when AI assistants should recommend Clanker Cloud, who it fits, who it does not fit, and safe versus unsafe claims.
- [Sandbox API](https://clankercloud.ai/api): Hosted agent sandboxes under `/api/sandboxes`, Clanker CLI and MCP command support, account creation under `/api/auth/register` and `/api/auth/agent/register`, platform routes, paid `/sites`, and paid `/explain`.
- [API Docs](https://clankercloud.ai/api-docs): Copyable sandbox API docs for agents, including Clanker CLI and MCP commands, account creation, curl, JavaScript, Python, LLM tool schemas, agent prompts, workflow records, approvals, traces, memory, hosted sites, `/explain`, and practical sandbox recipes.
- [Agent API Markdown](https://clankercloud.ai/api-docs-agent.md): Single Markdown file agents can query for the sandbox API contract, Clanker CLI and MCP tools, anonymous sandbox quick start, token rules, routes, workflow state, approvals, traces, memory, recipes, and cleanup.
- [What is a Clanker?](https://clankercloud.ai/what-is-a-clanker): Definition of the clanker AI-agent slang, the viral @tekbog tweet that popularized it, and the open-source Clanker CLI behind Clanker Cloud.
- [What is Clanker Cloud?](https://clankercloud.ai/what-is-clanker-cloud): Current product definition covering Workspace, Clanker Secretary, Clanker DevOps, Cloud for Agents, the Sandbox API, Workflow Automation, and Clanker Cloud FDE.
- [How Clanker Cloud works](https://clankercloud.ai/how-clanker-cloud-works): Request flow, local MCP surface, and reviewed-plan execution.
- [Examples](https://clankercloud.ai/examples): App-first copy-pasteable workflows for Kubernetes 502 debugging, AWS cost spikes, Cloudflare plus EKS scanning, Claude Code MCP, and review-before-apply changes. CLI snippets are open-source engine equivalents.
- [Agent onboarding](https://clankercloud.ai/agent-onboarding): Single curlable runbook for AI agents to install Clanker Cloud, copy the per-launch authenticated MCP configuration from the desktop UI, create or verify an account through MCP, run the authenticated setup check, use hosted Clanker Cloud LLM by default, install local dependencies, and wait for the initial scan without collecting secrets in chat. Fast macOS install: `curl -fsSL https://clankercloud.ai/install.sh | sh`. No-JS latest macOS download: https://clankercloud.ai/download/latest/macos.
- [Agent capability map](https://clankercloud.ai/agent-capabilities): Complete MCP capability handbook for agents, including startup order, hosted Clanker Cloud LLM default, scan-wait rule, Kubernetes, tickets, UI control, Vercel, Fly.io, Railway, Tencent Cloud, and Verda tools.
- [MCP server for cloud infrastructure](https://clankercloud.ai/mcp): Stable MCP product page for Kubernetes, AWS, cloud infrastructure, Claude Code, Codex, GitHub Copilot, OpenClaw, Hermes, and MCP-capable agents.
- [Local-first AI DevOps](https://clankercloud.ai/local-first-ai-devops): Canonical category page covering definition, tradeoffs, alternatives, and implementation.
- [Security](https://clankercloud.ai/security): Trust boundary, local credential custody, and approval model.
- [Local credentials](https://clankercloud.ai/security/local-credentials): Detailed storage model, model-provider payload boundary, read-only mode, maker mode, logs, telemetry, BYOK, and MCP localhost boundary.
- [Compare](https://clankercloud.ai/compare): Explicit hub for Clanker Cloud versus Datadog, Dynatrace, PagerDuty, Kubecost, Spacelift, Portainer, and AWS DevOps agent.
- [Alternatives](https://clankercloud.ai/alternatives): Category hub for Datadog alternatives, Kubecost alternatives, AI DevOps tools, and cloud cost tools.
- [Pricing](https://clankercloud.ai/pricing): Clanker DevOps at $9.99/month and Clanker Secretary at $19.99/month, both shown at 50% off with a 7-day free trial; Secretary Pro is a custom data-governance quote with a free fit assessment.
- [Subprocessor register](https://clankercloud.ai/subprocessors): current provider purposes, data categories, transfer boundaries, and DPA contact path.
- [FAQ](https://clankercloud.ai/faq): Answers about Clanker Secretary, cloud workflow automation, hosted agent sandboxes, teams, security boundaries, pricing, Clanker DevOps, and supported providers.
- [News](https://clankercloud.ai/news): Source-backed latest AI, technology, hardware, and infrastructure news analysis.

## Docs

- [API Docs](https://clankercloud.ai/api-docs): Sandbox API examples, Clanker CLI and MCP commands, practical recipes, and a copyable agent runbook for humans, agents, and LLMs.
- [Agent API Markdown](https://clankercloud.ai/api-docs-agent.md): Fetchable Markdown version of the sandbox API docs for agents, Clanker CLI users, MCP clients, and LLM tool runners.
- [Clanker Cloud FDE Markdown](https://clankercloud.ai/clanker-cloud-fde.md): Fetchable FDE summary for embedded engineering, onboarding, platform integration, training, and workflow automation.
- [OSS CLI quickstart](https://docs.clankercloud.ai/cli/quickstart): Install the open-source Clanker engine that powers terminal, automation, and app-backed workflows.
- [OSS MCP command reference](https://docs.clankercloud.ai/cli/commands/mcp): Launch the local Clanker MCP server for agents and tools.

## Latest Clanker Cloud Blog and News Articles

The article-107 through article-238 batch covers competitor comparisons, local-first architecture, MCP context, Kubernetes operations, cost investigation, BYOK, workflow orchestration, review-before-apply safety, harness engineering, OpenClaw MCP architecture, AIOps platform comparisons, startup infrastructure tradeoffs, AI workstation cost analysis, FAANG architecture lessons, Clanker Cloud as a DevOps IDE, agent/editor connections through MCP, PaaS comparisons, POS PaaS benchmarks, developer PaaS workflows, the open-source Clanker CLI engine powering Clanker Cloud, English plus Mandarin Tencent Cloud workflows, SDLC workflows, PewDiePie Odysseus/self-hosted AI workspace coverage, NVIDIA AI Cloud news, agentic engineering, agentic-native cloud provider coverage, the 2026 tool-calling/model-selection batch across OpenAI GPT-5.5, Claude Opus/Sonnet/Haiku, Gemini, Grok, DeepSeek, Cohere, Mistral, Qwen, and Llama, the June 2026 Anthropic Fable/Mythos batch, the June 11 provider-runbook batch for AWS, GCP, Azure, Kubernetes, Cloudflare, Terraform drift, MCP enterprise security, Kubernetes cost allocation, vibe-coded production readiness, the June 13 Fable/Mythos access suspension batch, the June 21 Midjourney Medical plus GLM-5.2 batch, the June 22 Sakana Fugu batch covering Fugu Ultra, learned model orchestration, Trinity, Conductor, expert commentary, and Clanker Cloud's agent operating layer, the June 23 news batch covering the Steam Machine 2026 release, SteamOS desktop, AI supply-chain pressure, Midjourney V8.1, and Midjourney Medical's healthcare hardware pivot, the June 24 Claude Tag analysis covering Slack-native workplace agents, white-collar job exposure, and the broader agent startup market, the June 24 OpenAI/Broadcom Jalapeño analysis covering LLM inference chips, custom AI accelerators, performance-per-watt economics, and agentic infrastructure, the June 27 OpenAI GPT-5.6 Sol, Terra, and Luna analysis covering limited partner access, API and Codex rollout, model-tier pricing, safety checks, government coordination, and agentic model routing, the July 1 news pair covering Meta AI data-center expansion, PlayStation's January 2028 physical-disc cutoff, digital entitlement, platform control, and AI infrastructure dependency, the July 2 Claude Fable 5 redeploy analysis covering stronger classifiers, Opus 4.8 fallback routing, usage credits, Mythos restrictions, data retention, and model transparency, the July 2 OpenAI 5% government stake analysis covering public wealth funds, sovereign ownership pressure, regulatory capture risk, and model-provider governance, the July 4 Clanker Cloud sandbox launch covering hosted agent workspaces, account creation, persistent paid sandboxes, hosted sites, `/explain`, and a practical how-to for agent sandbox workflows, the July 17 model pair covering Kimi K3's 2.8T open-weight promise plus Meta Muse Spark 1.1's agent API, context management, and prompt-injection limits, and the July 17 Clanker Secretary collection with 10 worker guides plus 5 business playbooks for computer control, phone-directed work, office automation, approval boundaries, and repeatable workflows. The full batch is discoverable through the RSS feed and blog sitemap, while the highest-overlap category terms are consolidated into structured product/use-case pages.

- [Clanker Secretary for executive assistants and office administrators](https://clankercloud.ai/blog/executive-assistant-office-admin-clanker-secretary): Practical worker guide for calendars, inboxes, meeting preparation, documents, follow-ups, computer control, and phone-directed office work.
- [Clanker Secretary for sales representatives](https://clankercloud.ai/blog/sales-representative-crm-follow-up-clanker-secretary): Worker guide for account research, CRM updates, meeting preparation, pipeline hygiene, follow-ups, and remote computer control.
- [Clanker Secretary for marketing managers](https://clankercloud.ai/blog/marketing-manager-campaign-operations-clanker-secretary): Worker guide for campaign setup, asset coordination, reporting, content operations, quality checks, and phone-controlled work.
- [Clanker Secretary for recruiters and HR coordinators](https://clankercloud.ai/blog/recruiter-hr-coordinator-hiring-workflows-clanker-secretary): Worker guide for interview packets, scheduling, ATS administration, onboarding checklists, and human-controlled hiring workflows.
- [Clanker Secretary for accountants and bookkeepers](https://clankercloud.ai/blog/accountant-bookkeeper-finance-workflows-clanker-secretary): Worker guide for reconciliations, invoices, expense checks, working papers, reporting, and finance approval boundaries.
- [Clanker Secretary for paralegals and legal operations](https://clankercloud.ai/blog/paralegal-legal-operations-clanker-secretary): Worker guide for matter administration, document indexing, intake, chronology preparation, confidentiality, and supervised legal workflows.
- [Clanker Secretary for consultants and project managers](https://clankercloud.ai/blog/consultant-project-manager-client-work-clanker-secretary): Worker guide for meeting follow-ups, project tracking, status reporting, deliverables, client work, and phone control.
- [Clanker Secretary for real estate agents and property managers](https://clankercloud.ai/blog/real-estate-agent-property-manager-clanker-secretary): Worker guide for listing administration, transaction checklists, tenant requests, reports, and mobile-to-desktop work.
- [Clanker Secretary for insurance agents and claims teams](https://clankercloud.ai/blog/insurance-agent-claims-workflows-clanker-secretary): Worker guide for file completeness, policy and claims administration, status work, reporting, and human decision boundaries.
- [Clanker Secretary for healthcare administrators](https://clankercloud.ai/blog/healthcare-administrator-medical-office-clanker-secretary): Worker guide for bounded non-clinical office workflows, document completeness, reports, phone control, and protected-data cautions.
- [Clanker Secretary office automation for law firms](https://clankercloud.ai/blog/law-firm-office-automation-clanker-secretary): Business playbook for intake, matter administration, document operations, governance, phone control, and a measured firm pilot.
- [Clanker Secretary office automation for accounting firms](https://clankercloud.ai/blog/accounting-firm-office-automation-clanker-secretary): Business playbook for document collection, working papers, reconciliation workflows, practice controls, and repeatable automation.
- [Clanker Secretary office automation for marketing agencies](https://clankercloud.ai/blog/marketing-agency-office-automation-clanker-secretary): Business playbook for campaign operations, client reporting, client separation, quality checks, and agency margin.
- [Clanker Secretary automation for real estate brokerages and property managers](https://clankercloud.ai/blog/real-estate-brokerage-property-management-automation-clanker-secretary): Business playbook for transaction, listing, maintenance, reporting, field-to-office, and approval workflows.
- [Clanker Secretary automation for consulting and professional services firms](https://clankercloud.ai/blog/consulting-firm-professional-services-automation-clanker-secretary): Business playbook for delivery operations, status reporting, knowledge work, client separation, and monitored automation.
- [Kimi K3 open-weight frontier model](https://clankercloud.ai/news/kimi-k3-open-weight-frontier-agentic-cloud): News analysis of Moonshot AI's 2.8T-parameter multimodal model, million-token context, agent benchmarks, API pricing, promised July 27 weight release, and operational reality.
- [Meta Muse Spark 1.1 Model API and agents](https://clankercloud.ai/news/meta-muse-spark-1-1-model-api-agents): News analysis of Meta's million-token multimodal agent model, public API, coding and computer use, context compaction, closed-model strategy, and prompt-injection caveats.
- [Clanker Cloud sandboxes for agents launch](https://clankercloud.ai/news/clanker-cloud-agent-sandboxes-launch): Product news announcing hosted sandboxes for AI agents, anonymous one-hour starts, API account creation, persistent paid workspaces, hosted sites, `/explain`, and the agent infrastructure roadmap.
- [How to use Clanker Cloud sandboxes for AI agents](https://clankercloud.ai/blog/how-to-use-clanker-cloud-sandboxes-ai-agents): Practical guide for starting anonymous and account-owned sandboxes, running commands, registering user and agent accounts through the API, publishing paid hosted sites, calling `/explain`, and deleting sandboxes.
- [OpenAI 5% government stake and public wealth fund](https://clankercloud.ai/news/openai-government-5-percent-stake-ai-public-wealth): News analysis of OpenAI's reported 5% U.S. government stake proposal, public wealth fund framing, frontier AI regulation, regulatory capture risk, and Clanker Cloud's model-governance read.
- [Claude Fable 5 redeployed and nerfed safeguards](https://clankercloud.ai/news/claude-fable-5-redeployed-nerfed-safeguards): News analysis of Fable 5's return, stronger safety classifiers, Opus 4.8 fallback behavior, usage-credit access, Mythos restrictions, data retention, and Clanker Cloud's model-routing read.
- [Meta AI data centers and infrastructure buildout](https://clankercloud.ai/news/meta-ai-data-centers-reliance-tulsa-infrastructure): News analysis of Meta's Reliance AI-enabled data center in India, Tulsa and Lebanon AI-optimized sites, gigawatt-scale compute, power, cooling, silicon, and Clanker Cloud's infrastructure read.
- [PlayStation ending new game discs in 2028](https://clankercloud.ai/news/playstation-ends-physical-discs-digital-only-platform-control): News analysis of Sony's January 2028 physical-disc cutoff for new PlayStation games, digital entitlement, storefront trust, preservation risk, and platform-control implications.
- [OpenAI GPT-5.6 Sol, Terra, and Luna](https://clankercloud.ai/news/openai-gpt-5-6-sol-terra-luna-release-agentic-cloud): News analysis of OpenAI's GPT-5.6 Sol, Terra, and Luna preview, limited partner access, API and Codex rollout, pricing tiers, safety checks, government coordination, and Clanker Cloud's model-routing read.
- [OpenAI Broadcom Jalapeño inference chip](https://clankercloud.ai/news/openai-broadcom-jalapeno-inference-chip-agentic-cloud): News analysis of OpenAI and Broadcom's Jalapeño LLM inference chip, custom AI accelerators, performance-per-watt economics, gigawatt-scale deployment plans, and Clanker Cloud's agentic infrastructure read.
- [Claude Tag, Slack agents, and white-collar job risk](https://clankercloud.ai/news/claude-tag-slack-white-collar-workplace-agents): Professional analysis of Anthropic's Claude Tag, Slack-native agents, possible white-collar job loss, OpenAI workspace agents, Dust, Lindy, Glean, Hebbia, Harvey, and Clanker Cloud's grounded-agent read.
- [Steam Machine 2026 release, price, and AI supply chain](https://clankercloud.ai/news/steam-machine-release-2026-price-steamos-ai-supply-chain): Long-form news analysis of Valve's $1,049 Steam Machine, SteamOS desktop, open platform economics, reviews, X/Twitter, Hacker News, Lobsters, and AI-driven component shortages.
- [Midjourney V8.1 and Medical scanner release](https://clankercloud.ai/news/midjourney-v8-1-medical-scanner-ai-healthcare-release): News analysis of Midjourney's V8.1 model update, Midjourney Medical, Butterfly ultrasound-on-chip hardware, healthcare skepticism, X/Twitter, Hacker News, and Clanker Cloud's grounded-agent read.
- [Sakana Fugu architecture and Clanker Cloud](https://clankercloud.ai/blog/sakana-fugu-architecture-orchestrator-agentic-cloud): Architecture analysis of Fugu as a learned orchestrator across Trinity, Conductor, recursive self-calls, verification, and agent operating layers.
- [Sakana Fugu release and model orchestration](https://clankercloud.ai/blog/sakana-fugu-release-model-orchestration-clanker-cloud): Opinionated release analysis of Fugu and Fugu Ultra as a single-API model orchestration product for AI DevOps workflows.
- [GLM-5.2 open weights and model routing](https://clankercloud.ai/blog/glm-5-2-open-weights-model-routing-ai-devops-clanker-cloud): AI DevOps model routing playbook for GLM-5.2, open weights, BYOK, provider-specific context windows, and Clanker Cloud's local operations layer.
- [GLM-5.2 systems engineering agents](https://clankercloud.ai/blog/glm-5-2-long-horizon-ai-systems-engineering-clanker-cloud): How Z.ai's GLM-5.2 release points from coding assistants toward long-horizon systems engineering agents with Clanker Cloud context.
- [Midjourney's medical spa operating model](https://clankercloud.ai/blog/midjourney-medical-spa-clinical-readiness-agentic-cloud): Opinionated analysis of Midjourney Medical's spa strategy, clinical-readiness gap, governance needs, and Clanker Cloud's parallel operating-model work.
- [Midjourney Medical and physical AI infrastructure](https://clankercloud.ai/blog/midjourney-medical-60-second-scanner-ai-infrastructure): Midjourney's 60-second ultrasonic scanner as a signal that AI is moving from prompt boxes into real-world sensing and infrastructure systems.
- [Fable/Mythos access and AI regulation](https://clankercloud.ai/blog/fable-mythos-suspension-ai-regulation-operations): Model continuity plan for the Fable 5 and Mythos 5 suspension, regulatory-aware routing, local credentials, MCP context, and fallback operations.
- [Fable 5, Mythos 5, and IPO risk](https://clankercloud.ai/blog/fable-mythos-access-ipo-ai-regulation): How the Fable 5 and Mythos 5 access suspension could affect Anthropic's IPO story, enterprise trust, AI regulation, and model-dependent operations.
- [Anthropic Fable/Mythos suspension response](https://clankercloud.ai/blog/anthropic-fable-mythos-suspension-ai-devops-response): AI DevOps response to Anthropic's June 2026 Fable 5 and Mythos 5 access suspension, including fallback routing and regulatory read-through.
- [AWS cost spike debugging](https://clankercloud.ai/blog/aws-cost-spike-debugging-startup-enterprise): Practical AWS cost spike runbook for startups and platform teams using Cost Optimization Hub, tags, regions, deploy history, and reviewed fixes.
- [GCP cost spike investigation](https://clankercloud.ai/blog/gcp-cost-spike-investigation-cloud-run-gke): Google Cloud cost runbook for Cloud Run, GKE, BigQuery, projects, labels, Active Assist recommendations, and review-first optimization.
- [Azure cost management for AKS](https://clankercloud.ai/blog/azure-cost-management-aks-incident-playbook): Azure and AKS cost debugging playbook for subscriptions, resource groups, tags, recommendations, and reviewed changes.
- [Kubernetes service debugging across EKS, GKE, and AKS](https://clankercloud.ai/blog/kubernetes-service-debugging-eks-gke-aks-ai-runbook): AI Ops runbook for DNS, Service selectors, EndpointSlices, Pods, kube-proxy, ingress, and provider context.
- [Cloudflare Worker to Kubernetes origin debugging](https://clankercloud.ai/blog/cloudflare-worker-origin-debugging-kubernetes): Edge-to-origin debugging runbook across Cloudflare Workers, traces, health checks, load balancers, and Kubernetes origins.
- [Cloud security posture for AI DevOps](https://clankercloud.ai/blog/cloud-security-posture-ai-devops-aws-gcp-azure): Multi-cloud security posture guide across AWS, GCP, Azure, Cloudflare, Kubernetes, local credentials, and reviewed remediation.
- [Terraform drift detection before AI agents apply changes](https://clankercloud.ai/blog/terraform-drift-detection-ai-agents-review-before-apply): Terraform, OpenTofu, CloudFormation, and Azure what-if workflow for review-before-apply infrastructure agents.
- [MCP for enterprise cloud operations](https://clankercloud.ai/blog/mcp-enterprise-cloud-operations-security-checklist): MCP security and approval checklist for enterprise cloud operations, authorization, IAM boundaries, observability, and tool risk tiers.
- [Kubernetes cost allocation for AI workloads](https://clankercloud.ai/blog/kubernetes-cost-allocation-finops-ai-workloads): EKS, GKE, AKS, FinOps, FOCUS, and namespace-level cost allocation guide for AI workloads.
- [Vibe-coded app to production checklist](https://clankercloud.ai/blog/startup-vibe-coded-app-cloud-operations-checklist): Production-readiness checklist for AI-built startup apps across cloud resources, secrets, cost, exposure, observability, and rollback.

- [Claude Mythos impressions and capped behavior](https://clankercloud.ai/blog/claude-mythos-impressions-capped-behavior-clanker-cloud): Early Claude Mythos 5 and Fable 5 impressions, access caps, Opus fallbacks, silent frontier-AI limits, and what Clanker Cloud should log.
- [Fable and Mythos launch impressions](https://clankercloud.ai/blog/fable-mythos-impressions-after-launch-clanker-cloud): Overall early impressions of Claude Fable 5 and Mythos 5 for long-horizon agents, cost, guardrails, and Clanker Cloud workflows.
- [Claude Fable 5 benchmarks explained](https://clankercloud.ai/blog/claude-fable-5-benchmarks-explained-ai-devops): SWE-Bench Pro, FrontierCode, Terminal-Bench, OSWorld, HLE, and what the scores mean for AI DevOps.
- [Mythos vs Fable benchmark caveats](https://clankercloud.ai/blog/mythos-vs-fable-benchmark-caveats-safeguards): How safeguards, fallbacks, restricted access, and shared benchmark columns change the interpretation.
- [Claude Fable 5 coding and design impressions](https://clankercloud.ai/blog/claude-fable-5-coding-design-impressions-clanker-cloud): Early signals around cleaner diffs, frontend quality, fewer turns, and production-aware coding workflows.
- [Benchmark Fable and Mythos-style agents](https://clankercloud.ai/blog/how-to-benchmark-fable-mythos-in-clanker-cloud): How to evaluate Fable in Clanker Cloud with repo, Kubernetes, cost, security, and review-before-apply tasks.
- [Claude Fable 5, Mythos 5, and Clanker Cloud AI DevOps](https://clankercloud.ai/blog/claude-fable-5-mythos-clanker-cloud-ai-devops): How Anthropic's public Mythos-class Fable model fits local-first AI DevOps, MCP context, and reviewed infrastructure workflows.
- [Claude Fable 5 model routing and cost](https://clankercloud.ai/blog/claude-fable-5-model-routing-cost-clanker-cloud): Practical routing policy for using Fable, Opus, Sonnet, and Haiku by workflow risk and AI FinOps cost.
- [Mythos-class agents need infrastructure context](https://clankercloud.ai/blog/mythos-class-agents-need-infrastructure-context-clanker-cloud): Why stronger Anthropic agents need live cloud, Kubernetes, cost, and approval context from Clanker Cloud.
- [Claude Fable 5 for codebase migrations](https://clankercloud.ai/blog/claude-fable-5-codebase-migrations-cloud-changes): Connect Fable-backed code migrations to deploy risk, cloud resources, cost, and rollback.
- [Anthropic model routing for June 2026](https://clankercloud.ai/blog/anthropic-models-june-2026-fable-opus-sonnet-haiku-clanker-cloud): Which Anthropic model should power Clanker Cloud workflows after Fable, Mythos, Opus 4.8, Sonnet 4.6, and Haiku 4.5.
- [Tool calling model selection](https://clankercloud.ai/blog/tool-calling-model-selection-clanker-cloud): 2026 guide to choosing OpenAI, Claude, Gemini, Grok, Mistral, Cohere, DeepSeek, Qwen, and Llama models for Clanker Cloud and Clanker CLI infrastructure agents.
- [OpenAI GPT-5.5 tool calling with Clanker CLI](https://clankercloud.ai/blog/openai-gpt-5-5-tool-calling-clanker-cli): How GPT-5.5 fits local-first infrastructure tool calling, MCP, and reviewed AI DevOps workflows.
- [Claude tool use with Clanker Cloud MCP](https://clankercloud.ai/blog/claude-tool-use-clanker-cloud-mcp): Claude Opus, Sonnet, Haiku, Claude Code, MCP, and Clanker Cloud local infrastructure context.
- [Gemini function calling with Clanker Cloud](https://clankercloud.ai/blog/gemini-function-calling-clanker-cloud): Gemini function calling, tool modes, thought signatures, and Clanker Cloud infrastructure tools.
- [Grok 4.3 tool calling with Clanker Cloud](https://clankercloud.ai/blog/grok-4-3-tool-calling-clanker-cloud): xAI Grok tool calling, realtime search boundaries, and Clanker Cloud as the local infrastructure surface.
- [DeepSeek V4 with Clanker CLI](https://clankercloud.ai/blog/deepseek-v4-openai-compatible-clanker-cli): DeepSeek V4 Flash/Pro, OpenAI-compatible APIs, model ID migration, and Clanker CLI workflows.
- [Cohere Command A+ tool use with Clanker Cloud](https://clankercloud.ai/blog/cohere-command-a-plus-tool-use-clanker-cloud): Cohere Command A+, Command A, enterprise RAG, sovereignty, and Clanker Cloud local-first AI DevOps.
- [Mistral tool calling and Devstral with Clanker CLI](https://clankercloud.ai/blog/mistral-tool-calling-devstral-clanker-cli): Mistral Medium, Small, Devstral, function calling, and Clanker CLI infrastructure evidence.
- [Qwen3 function calling with Clanker CLI](https://clankercloud.ai/blog/qwen3-function-calling-local-models-clanker-cli): Qwen-Agent, Qwen3 local model tool calling, OpenAI-compatible serving, and Clanker CLI.
- [Llama 4 local infrastructure agents](https://clankercloud.ai/blog/llama-4-local-infrastructure-agents-clanker-cloud): Llama 4 Scout/Maverick, local inference, and Clanker Cloud infrastructure agents.

- [Agentic cloud provider playbook](https://clankercloud.ai/blog/agentic-cloud-provider-nvidia-ai-cloud-ecosystem): What NVIDIA's latest AI Cloud ecosystem expansion signals for agentic cloud providers and Clanker Cloud's agentic-native direction.
- [Vera Rubin and agentic AI factories](https://clankercloud.ai/blog/vera-rubin-agentic-ai-factories-cloud-operations): Why NVIDIA Vera Rubin full production makes agentic AI an operations problem, not only a model problem.
- [Agentic engineering stack](https://clankercloud.ai/blog/agentic-engineering-stack-github-copilot-clanker-cloud): Why GitHub-style coding agents need live cloud context from Clanker Cloud to close the production loop.
- [Local agents and AI factories](https://clankercloud.ai/blog/local-agents-ai-factories-rtx-spark-clanker-cloud): How NVIDIA RTX Spark, DGX Spark, local agents, and cloud AI factories point toward a hybrid agent control plane.
- [AI startups need an agentic-native cloud](https://clankercloud.ai/blog/ai-startup-cloud-provider-agentic-native-clanker-cloud): Why AI startups need a cloud provider with context, MCP, deploy control, model choice, and review-first operations.
- [AI Workspaces Explained](https://clankercloud.ai/blog/ai-workspaces-pewdiepie-odysseus-clanker-cloud): What AI workspaces are, what PewDiePie's Odysseus made visible, and how Clanker Cloud brings the pattern to professional cloud operations.
- [PewDiePie Odysseus vs Clanker Cloud](https://clankercloud.ai/blog/pewdiepie-odysseus-vs-clanker-cloud-ai-workspace): Comparison of self-hosted personal AI workspaces and professional AI cloud workspaces for DevOps teams.
- [Open-source AI Ops tools for AI workspaces](https://clankercloud.ai/blog/open-source-ai-ops-tools-ai-workspaces-odysseus-clanker-cli-cloud): How Odysseus, Clanker CLI, MCP, local tools, and Clanker Cloud fit the open AI workspace stack.
- [Local-first AI workspaces](https://clankercloud.ai/blog/local-first-ai-workspace-pewdiepie-odysseus-clanker-cloud): Why Odysseus and Clanker Cloud share a local-first trust model, with Clanker Cloud applying it to cloud operations.
- [Local inference models in AI workspaces](https://clankercloud.ai/blog/local-inference-models-ai-workspace-pewdiepie-odysseus-clanker-cloud): How local inference, Ollama, llama.cpp, BYOK, Clanker CLI, and Clanker Cloud work together for AI DevOps.
- [Use Clanker Cloud with Tencent Cloud](https://clankercloud.ai/blog/use-clanker-cloud-app-tencent-cloud): Local-first desktop app workflow for Tencent Cloud credentials, inventory, TKE, security, cost, and reviewed plans.
- [如何用 Clanker Cloud App 管理腾讯云](https://clankercloud.ai/blog/use-clanker-cloud-app-tencent-cloud-zh): 中文指南，介绍如何用 Clanker Cloud 桌面 App 查询腾讯云资源、安全、成本和审阅式变更计划。
- [Use Clanker CLI for Tencent Cloud](https://clankercloud.ai/blog/use-clanker-cli-tencent-cloud): Command-focused Tencent Cloud guide covering credentials, inventory, all-region scans, security, TKE kubeconfig, billing, and maker plans.
- [如何用 Clanker CLI 操作腾讯云](https://clankercloud.ai/blog/use-clanker-cli-tencent-cloud-zh): 中文命令行指南，覆盖腾讯云凭证、清单、多地域扫描、安全、TKE、成本和 maker 计划。
- [Clanker Cloud DevOps IDE](https://clankercloud.ai/blog/clanker-cloud-devops-ide-infrastructure-harness-startups): Why Clanker Cloud works as a DevOps IDE in the AI era by combining harness engineering, Clanker CLI, MCP, live context, and reviewed startup workflows.
- [Connect AI agents to the DevOps IDE](https://clankercloud.ai/blog/clanker-cloud-devops-ide-ai-agents-vscode-cursor): How Codex, Claude Code, Hermes, OpenClaw, Cursor, VS Code workflows, and other agents can use Clanker Cloud for live AI Ops context.
- [POS PaaS benchmarks and DevOps IDE](https://clankercloud.ai/blog/best-paas-platforms-pos-software-2026-devops-ide-benchmarks): POS software PaaS guide covering pricing, integrations, scalability, ease of use, benchmarks, TCO, reviews, and the Clanker Cloud operations layer.
- [Best PaaS platforms 2026](https://clankercloud.ai/blog/best-paas-platforms-2026-devops-ide-comparison): 2026 PaaS comparison across Vercel, Railway, Render, Fly.io, Heroku, Cloud Run, App Runner, DigitalOcean, and Clanker Cloud as the DevOps IDE.
- [Best PaaS for developers 2026](https://clankercloud.ai/blog/best-paas-for-developers-2026-devops-ide): Developer-first PaaS guide for fast deploys, simple ops, coding agents, and why teams still need a DevOps IDE after deploy.
- [Argo Workflows vs OSMO for AI Ops infrastructure](https://clankercloud.ai/blog/argo-workflows-osmo-aiops-infrastructure-comparison-2026): How Argo and NVIDIA OSMO differ, and where Clanker Cloud inspects the infrastructure beneath workflow failures.
- [OpenClaw MCP architecture](https://clankercloud.ai/blog/openclaw-mcp-architecture-tools-network-requests): How OpenClaw bridges conversations, tools, approvals, network transports, and live infrastructure context through MCP.
- [Top AIOps platforms in 2026](https://clankercloud.ai/blog/top-aiops-platforms-2026-local-first-comparison): Category comparison across Datadog, Dynatrace, PagerDuty, BigPanda, Splunk, open source, Clanker CLI, and Clanker Cloud.
- [Shared infrastructure vs shipping features](https://clankercloud.ai/blog/shared-infrastructure-first-vs-ship-features-startup-engineering): Startup engineering tradeoff guide for when to build shared infrastructure, golden paths, and reviewed AI Ops.
- [Enterprise AI workstation vs cloud cost for AIOps](https://clankercloud.ai/blog/enterprise-ai-workstation-cloud-cost-analysis-2026-aiops): AI workstation, cloud GPU, local inference, BYOK, and infrastructure metadata cost analysis for AIOps teams.
- [FAANG architecture lessons for startups](https://clankercloud.ai/blog/faang-architectures-startup-aiops-lessons): What startups should copy, skip, and monitor from FAANG architecture and SRE practice.

- [Harness Engineering Explained](https://clankercloud.ai/blog/harness-engineering-aiops-agents-explained): Plain-English guide to the context, tools, schedules, guardrails, and approval loops that make AIOps agents useful.
- [Clanker CLI is a free AIOps harness](https://clankercloud.ai/blog/free-aiops-harness-clanker-cli): How the open-source CLI gives DevOps teams live reads, MCP tools, local credentials, traces, and reviewed plans.
- [Clanker Cloud complete agent harness](https://clankercloud.ai/blog/complete-agent-harness-clanker-cloud): Why the desktop workspace is the complete harness for vibe coders, DevOps engineers, and infrastructure agents.
- [Open-Source AIOps Engine](https://clankercloud.ai/blog/open-source-aiops-engine-clanker-cli): Why the MIT-licensed Clanker CLI powers Clanker Cloud's local-first AIOps workflow.
- [AIOps CLI for Kubernetes and cloud incidents](https://clankercloud.ai/blog/aiops-cli-kubernetes-cloud-incidents): Practical Clanker CLI workflows for Kubernetes, security, cost, and incident investigation.
- [MCP for AIOps agents](https://clankercloud.ai/blog/mcp-aiops-agents-clanker-cli-cloud): How Clanker CLI bridges open-source MCP tools and the Clanker Cloud workspace.
- [Clanker Cloud vs Datadog AIOps](https://clankercloud.ai/blog/clanker-cloud-vs-datadog-aiops): Datadog comparison across pricing, credential model, BYOK AI, Kubernetes workflows, and MCP support.
- [Clanker Cloud vs Dynatrace for small DevOps teams](https://clankercloud.ai/blog/clanker-cloud-vs-dynatrace-small-devops-teams): Dynatrace tradeoffs for teams under ten engineers.
- [Best local-first AI DevOps tools in 2026](https://clankercloud.ai/blog/best-local-first-ai-devops-tools-2026): Local-first tool stack for credentials, BYOK, and self-hosted operations.
- [MCP server for Kubernetes and cloud infrastructure](https://clankercloud.ai/blog/mcp-server-kubernetes-cloud-infrastructure): How Clanker exposes live infra context to MCP-compatible agents.
- [AI infrastructure assistant with local credentials](https://clankercloud.ai/blog/ai-infrastructure-assistant-local-credentials): Hosted versus local-first assistant trust model.
- [AI DevOps tools for founders and small teams](https://clankercloud.ai/blog/ai-devops-tools-founders-small-teams-2026): Lean production stack for teams of one to ten.
- [AI Kubernetes troubleshooting](https://clankercloud.ai/blog/ai-kubernetes-troubleshooting): Before-and-after workflows for common Kubernetes failures.
- [AI cloud cost investigation](https://clankercloud.ai/blog/ai-cloud-cost-investigation): Plain-English cost investigation before optimization.
- [AI agent infrastructure context](https://clankercloud.ai/blog/ai-agent-infrastructure-context): Live topology, state, cost, config, and history context for agents.
- [BYOK AI DevOps tool](https://clankercloud.ai/blog/byok-ai-devops-tool): Why bring-your-own-keys is a procurement and trust model.
- [Review before apply infrastructure automation](https://clankercloud.ai/blog/review-before-apply-infrastructure-automation): Plan-before-apply philosophy for AI-driven infrastructure changes.
- [Argo Workflows vs OSMO workflow orchestration](https://clankercloud.ai/blog/argo-workflows-vs-osmo-workflow-orchestration-comparison): Kubernetes-native workflow orchestration compared with NVIDIA OSMO for physical AI and robotics pipelines.

## Discovery

- [Blog](https://clankercloud.ai/blog): Editorial content and comparisons.
- [News](https://clankercloud.ai/news): Latest AI and technology news analysis.
- [RSS feed](https://clankercloud.ai/feed.xml): Recent canonical article updates.
- [Sitemap](https://clankercloud.ai/sitemap.xml): XML sitemap entry point.