# Luxor Agent Cloud > Agent-readable map for Luxor Agent Cloud, a cloud infrastructure platform for software teams and AI coding agents. Luxor Agent Cloud provides three products: - **Sandbox** provides disposable full Linux VMs for coding agents, development workloads, and untrusted code. - **Runners** provide managed Linux and macOS runners for GitHub Actions workflows. - **Code Reviewer** reviews pull requests for high-signal bugs, security issues, and code-quality problems. For current pricing, limits, and security details, use the linked documentation. Actions: - Search the documentation: `GET https://luxor.ai/api/search?query=`. - Create an account: [Luxor Agent Cloud registration](https://app.tenki.cloud/auth/registration/). - Open an existing workspace: [Luxor Agent Cloud dashboard](https://app.tenki.cloud). ## Choose your goal - [Run code in an isolated full Linux VM](https://luxor.ai/docs/sandbox/quickstart.md): Install the CLI or SDK, authenticate, create a session, run a command, and terminate it. - [Move a GitHub Actions job](https://luxor.ai/docs/runners/quickstart.md): Create an account and connect the repository through the Luxor Agent Cloud GitHub App; then replace the workflow's `runs-on` value. - [Enable automated pull-request review](https://luxor.ai/docs/start-code-review.md): Create an account and install the separate Reviewer GitHub App; then open a pull request or mention `@luxor-reviewer`. - [Onboard a user with an AI agent](https://luxor.ai/skills/tenki-onboarding/SKILL.md): Install the CLI, run `luxor onboard`, open the approval URL, and let it wait for approval; then set up Sandbox, Runners, or Code Reviewer. ## Start here - [Luxor Agent Cloud homepage](https://luxor.ai/index.md): Overview of Luxor Agent Cloud products and use cases. - [Luxor Agent Cloud documentation](https://luxor.ai/docs.md): Overview of Runners, Sandbox, and Code Reviewer. - [Pricing](https://luxor.ai/docs/pricing.md): Current plans, rates, credits, quotas, and billing. ## Agent interfaces - [Luxor Agent Cloud Docs MCP metadata](https://luxor.ai/.well-known/mcp): Connection metadata for the documentation MCP server available at POST /mcp. - [Agent Skills catalog](https://luxor.ai/.well-known/agent-skills/index.json): Machine-readable catalog of Luxor Agent Cloud skills with integrity hashes. - [Luxor Agent Cloud documentation skill](https://luxor.ai/skills/tenki-cloud-docs/SKILL.md): Guidance for researching and answering questions about Luxor Agent Cloud. ## Sandbox - [Sandbox overview](https://luxor.ai/products/sandbox.md): Disposable full Linux VMs for coding agents, development workloads, and untrusted code. - [Sandbox concepts](https://luxor.ai/docs/sandbox/concepts.md): Understand the primitives behind Sandbox, including sessions, persistent volumes, snapshots, and templates, and when to reach for each. - [Sandbox sessions](https://luxor.ai/docs/sandbox/sessions.md): Create and drive Sandbox sessions covering lifecycle, command execution, file I/O, port exposure, and SSH access. - [Persistent Volumes](https://luxor.ai/docs/sandbox/volumes.md): Use Sandbox persistent volumes for durable build caches, package caches, large repositories, and datasets that survive session termination. - [Storage Mounts](https://luxor.ai/docs/sandbox/storage.md): Mount S3, R2, GCS, and S3-compatible buckets as folders in a sandbox. - [Snapshots](https://luxor.ai/docs/sandbox/snapshots.md): Capture a point-in-time VM snapshot of any Sandbox session, then restore it later as a new session. - [Templates](https://luxor.ai/docs/sandbox/templates.md): Define a typed template from a Git context, build it into a private digest-addressed image, and create sessions from it. - [Using secrets in a sandbox](https://luxor.ai/docs/sandbox/secrets.md): Give your sandbox the credentials it needs through requests, environment variables, or files. - [Tailscale](https://luxor.ai/docs/sandbox/tailscale.md): Join a sandbox to your tailnet and route internet traffic through an exit node. - [Sandbox recipes](https://luxor.ai/docs/sandbox/recipes.md): Task-oriented recipes for common Sandbox workflows, including running AI agents, checkpointing environments, caching dependencies, and executing untrusted code. - [Sandbox SDK reference](https://luxor.ai/docs/sandbox/sdk.md): Programmatic reference for Sandbox covering installation, authentication, client options, identity discovery, OpenCode integration, and error types. - [Run Chrome in a sandbox](https://luxor.ai/docs/sandbox/chrome.md): Headless or headed Chrome, CDP over a preview URL, noVNC viewer, and computer-use agent control. - [Sandbox CLI reference](https://luxor.ai/docs/sandbox/cli.md): Command reference for the `luxor` CLI covering authentication, session lifecycle, command execution, file transfer, port exposure, SSH, persistent volumes, and snapshots. - [Base Image](https://luxor.ai/docs/sandbox/base-image.md): What every Sandbox session ships with out of the box, and how to add your own tools. - [Crabbox on a sandbox](https://luxor.ai/docs/sandbox/crabbox.md): Keep a sandbox warm as an always-on test box: Crabbox syncs your working tree and runs your tests there. - [Harbor agent evaluations on Sandbox](https://luxor.ai/docs/sandbox/harbor.md): Run Harbor benchmarks like Terminal-Bench with every trial in its own sandbox, to evaluate coding agents and models in parallel. - [Sandbox troubleshooting](https://luxor.ai/docs/sandbox/troubleshooting.md): Diagnose common Sandbox issues including session creation hangs, missing volume data after restore, SSH errors, port exposure, and command timeouts. - [Anchor](https://luxor.ai/docs/sandbox/anchor.md): Start from the pre-built Anchor image and connect to a remote browser over CDP. - [Grok](https://luxor.ai/docs/sandbox/grok.md): Start from the pre-built Grok image and run xAI's coding agent headlessly. - [Pi](https://luxor.ai/docs/sandbox/pi.md): Start from the pre-built Pi image and run the open source Pi coding agent inside a per-session isolation boundary. - [Droid](https://luxor.ai/docs/sandbox/droid.md): Start from the pre-built Droid image and run Factory's coding agent headlessly with droid exec. - [Amp](https://luxor.ai/docs/sandbox/amp.md): Start from the pre-built Amp image and run Sourcegraph's coding agent headlessly. - [Prime Agent](https://luxor.ai/docs/sandbox/prime-agent.md): Start from the pre-built Prime Agent image and run Prime Intellect's coding and research harness headlessly. - [Hotdata](https://luxor.ai/docs/sandbox/hotdata.md): Start from the pre-built Hotdata image and give an agent its own database to load files into and query with SQL. - [TinyFish](https://luxor.ai/docs/sandbox/tinyfish.md): Start from the pre-built TinyFish image and drive TinyFish's web agent, remote browser, fetch, and search APIs. ## Runners - [Runners overview](https://luxor.ai/products/runners.md): Managed Linux and macOS runners for GitHub Actions workflows. - [Runner sizes and labels](https://luxor.ai/docs/runners/sizes.md): Standard runner configurations with CPU, memory, operating system, and Linux runs-on labels. - [Choosing a runner](https://luxor.ai/docs/runners/which-runner-should-i-use.md): Select the appropriate CPU and memory configuration for a workload. - [Linux x64 Runners](https://luxor.ai/docs/runners/x64-runners.md): Bare-metal Linux x64 GitHub Actions runners on 5th Gen AMD Ryzen processors, running the official GitHub Actions runner images. - [macOS Runners (M4 Pro)](https://luxor.ai/docs/runners/macos-runners.md): Apple Silicon (M4 Pro) macOS GitHub Actions runners, with five sizes, macOS 15 and 26 images, and monthly image refreshes. - [Runner caching](https://luxor.ai/docs/runners/caching.md): Hosted GitHub Actions cache and transparent dependency mirrors, enabled by default. - [Runner sticky disks](https://luxor.ai/docs/runners/sticky-disks.md): Persistent disks that keep dependency and compiler caches warm between jobs. - [Runner repository access](https://luxor.ai/docs/runners/repo-management.md): Choose which repositories Runners can access through their GitHub App. - [Networking & Egress](https://luxor.ai/docs/runners/networking.md): Egress traffic, IP ranges, and private connectivity options for CI workloads. - [Tailscale connections](https://luxor.ai/docs/runners/tailscale.md): Connect your tailnet once in workspace settings, then let runners and sandboxes join it without auth keys. - [Secrets & Signing](https://luxor.ai/docs/runners/secrets-and-signing.md): How Runners handle secrets, and how to import an Apple signing certificate into a macOS job. - [Limits, Concurrency & Cold Start](https://luxor.ai/docs/runners/limits-and-concurrency.md): Concurrency defaults per plan, cold start latency, queue handling, and billing behavior for Runners. - [Runner Images](https://luxor.ai/docs/runners/runner-images.md): What's preinstalled on Linux and macOS runner images, how often standard images are refreshed, and how custom images work on the Enterprise plan. - [Runners troubleshooting](https://luxor.ai/docs/runners/troubleshooting.md): Diagnose common runner setup, image, networking, and job failures. ## Code Reviewer - [Code Reviewer overview](https://luxor.ai/products/code-reviewer.md): AI pull-request review focused on high-signal bugs, security issues, and code-quality problems. - [Cursor Origin](https://luxor.ai/docs/reviewer/cursor-origin.md): Connect Code Reviewer to Cursor Origin and get reviews on your Origin pull requests. - [Custom Context](https://luxor.ai/docs/reviewer/custom-context.md): Code Reviewer reads your repository's AGENTS.md or CLAUDE.md to apply your team's rules, expectations, and preferences to every review. - [Review Anatomy](https://luxor.ai/docs/reviewer/review-anatomy.md): How a Code Reviewer review appears on your pull request, on GitHub and Cursor Origin, and how to read each part. - [Settings](https://luxor.ai/docs/reviewer/settings.md): Code Reviewer exposes a small set of configuration options to control when it runs and how much signal it produces, on GitHub and Cursor Origin. - [Reviewer repository access](https://luxor.ai/docs/reviewer/repo-management.md): Choose which repositories Code Reviewer can access through its GitHub App. - [Reviewer troubleshooting](https://luxor.ai/docs/reviewer/troubleshooting.md): Diagnose missing or failed pull-request reviews. ## Secrets - [Secrets overview](https://luxor.ai/docs/secrets.md): Store credentials in your workspace and choose how your applications use them. - [Manage secrets](https://luxor.ai/docs/secrets/manage-secrets.md): Create, inspect, rotate, and revoke workspace secrets. - [Request policies](https://luxor.ai/docs/secrets/request-policies.md): Choose which outgoing HTTPS requests can use your saved secrets. ## Accounts and GitHub - [Billing](https://luxor.ai/docs/account/billing.md): Understand how billing works in Luxor Agent Cloud for GitHub Actions runners. - [Free credits](https://luxor.ai/docs/account/free-credits.md): Learn how free credits work in Luxor Agent Cloud and how to use them to run GitHub Actions runners with better performance and cost improvements. - [API keys](https://luxor.ai/docs/account/api-keys.md): Create, store, rotate, and revoke Luxor Agent Cloud API keys for Sandbox SDKs, CLI automation, and CI. - [Members and settings](https://luxor.ai/docs/account/members-and-settings.md): Manage Luxor Agent Cloud members and workspace settings. Invite team members, assign roles, & configure your cloud environment for peak collaboration. - [Payments and plans](https://luxor.ai/docs/account/payment.md): Update your payment details, switch plans, review past invoices, and manage your subscription settings quickly and securely. - [Roles and permissions](https://luxor.ai/docs/account/roles-perms.md): Luxor Agent Cloud's role-based access system. Learn about Owner, Admin, and Standard roles and detailed permission breakdowns. - [GitHub organizations](https://luxor.ai/docs/github/gh-organization.md): How to connect and update your GitHub Organization with your Luxor Agent Cloud workspace. - [GitHub App access](https://luxor.ai/docs/github/gh-app-access-level.md): Review and understand what level of access each Luxor Agent Cloud GitHub App requires on your GitHub Organization. - [GitHub repository management](https://luxor.ai/docs/github/repo-management.md): Connect, update, or remove repositories available to a Luxor Agent Cloud workspace. ## Trust and policies - [Security and isolation](https://luxor.ai/docs/trust/security.md): Workload isolation, infrastructure security, and compliance posture. - [Acceptable Use Policy](https://luxor.ai/docs/acceptable-use-policy.md): Rules for permitted and prohibited use of Luxor Agent Cloud services. - [Editorial & Corrections Policy](https://luxor.ai/docs/editorial-policy.md): Luxor Agent Cloud's content sourcing, corrections, and update policy. - [Privacy Policy](https://luxor.ai/docs/privacy-policy.md): How Luxor Agent Cloud collects, uses, and protects personal data. - [Luxor Agent Cloud for Startups: Program Terms and Conditions](https://luxor.ai/docs/startups-program-terms.md): Eligibility and terms for the Luxor Agent Cloud for Startups program. - [Terms of Service](https://luxor.ai/docs/terms-of-service.md): Contractual terms governing use of Luxor Agent Cloud services. ## Blog posts - [Run TinyFish web agents from a Tenki Sandbox](https://luxor.ai/blog/tinyfish-tenki-sandbox.md): A pre-built Tenki image with the TinyFish CLI, SDK and Playwright client. Run web agents, fetch, search and remote browsers from an isolated VM. - [Give every agent task its own database with Hotdata and Tenki](https://luxor.ai/blog/hotdata-instant-databases-tenki-sandbox.md): Run your agent in a Tenki Sandbox and give each task its own Hotdata database to load files into and query with SQL, with the key kept out of images. - [Tenki Code Reviewer now runs on Cursor Origin](https://luxor.ai/blog/tenki-code-reviewer-cursor-origin.md): Install Tenki Code Reviewer from the Origin marketplace, and every pull request on the repos you select gets a summary, inline findings and a check. - [How to benchmark inference providers across coding-agent harnesses](https://luxor.ai/blog/benchmark-inference-providers-coding-agent-harnesses.md): Test any model or inference endpoint across coding-agent harnesses like Claude Code and OpenCode on Terminal-Bench, with every trial in a Tenki sandbox. - [Omnara agents can now run on Tenki Sandboxes](https://luxor.ai/blog/omnara-tenki-sandboxes.md): Omnara now supports Tenki Sandboxes as a machine pool provider. Add a Tenki API key and your Omnara agents get their own Linux VMs on demand. - [Naïve agent computers now run on Tenki Sandboxes](https://luxor.ai/blog/naive-tenki-sandboxes.md): Naïve's AI employees incorporate companies, launch campaigns and write code. Every task they run now gets its own fresh Tenki sandbox. - [Keep API tokens out of your sandbox](https://luxor.ai/blog/sandbox-secrets-transparent-injection.md): Tenki swaps a placeholder for your real API token outside the sandbox, so agent code can call an API without ever holding the credential. - [Sticky disks keep your GitHub Actions caches warm between jobs](https://luxor.ai/blog/sticky-disks-github-actions-cache.md): Tenki sticky disks keep dependency and compiler caches on a disk that follows the cache key, so jobs start warm without restoring a tarball. - [Connect a sandboxed AI agent to Anchor Browser over CDP](https://luxor.ai/blog/anchor-browser-tenki-sandbox.md): Run your agent in a Tenki Sandbox and drive an Anchor-managed browser with Playwright, with the API key kept out of images, templates and snapshots. - [How to benchmark GitHub Actions runners fairly](https://luxor.ai/blog/benchmarking-github-actions-runners.md): Compare CI runners on the same YAML, warm caches and repeated runs, measure cost per job instead of rate per minute, and see what our benchmark shows. - [A browser for AI agents: run Chrome in a disposable sandbox](https://luxor.ai/blog/browser-for-ai-agents-chrome-sandbox.md): Give an AI agent or computer-use model its own Chrome in a throwaway VM, driven over CDP with Playwright or Puppeteer and watchable through noVNC. - [Run your tests on a warm sandbox before you push](https://luxor.ai/blog/run-tests-before-you-push-warm-sandbox.md): Sync your uncommitted working tree to a warm Linux sandbox with Crabbox and run your test suite there, so your laptop and your coding agent stay free. - [Where to run code your AI agent wrote](https://luxor.ai/blog/run-untrusted-agent-code-sandbox.md): Your laptop, a container, a CI runner or a disposable VM? Where to run agent-generated code and untrusted MCP servers so a bad result stays contained. - [Sign and notarize macOS and iOS builds in GitHub Actions](https://luxor.ai/blog/sign-macos-ios-builds-github-actions.md): Import a .p12 into a temporary keychain, install profiles, notarize with notarytool or use fastlane match, without leaving signing keys on the runner. - [AGENTS.md and CLAUDE.md: shared instructions for AI coding agents](https://luxor.ai/blog/agents-md-shared-instructions-config.md): How Claude Code, Codex, Copilot and Cursor read AGENTS.md and CLAUDE.md, what to put in the file, and the anti-patterns that make it useless. - [GitHub Actions cost optimization: how to cut your CI bill](https://luxor.ai/blog/github-actions-cost-optimization.md): Reduce GitHub Actions costs with runner right-sizing, caching, concurrency controls, workflow audits, spend alerts and a lower per-minute rate. - [Self-hosted GitHub Actions runners: ARC, security, and cost](https://luxor.ai/blog/self-hosted-runners-arc-security-guide.md): How self-hosted GitHub Actions runners register and autoscale with ARC, an 8-point hardening checklist, and the operational work you take on. - [Copilot code review cost: AI credits plus Actions minutes](https://luxor.ai/blog/copilot-code-review-actions-minutes-cost.md): Copilot code review on private repos bills AI credits and GitHub Actions minutes. What each review costs, how to measure it, and which runners it can use. - [How we benchmark AI code reviewers: per-bug scoring on real PRs](https://luxor.ai/blog/benchmarking-ai-code-reviewers.md): How Tenki's code review benchmark scores seven AI reviewers bug by bug on 50 real pull requests, and what per-bug scoring does and doesn't show. - [The hidden CI tax of AI coding agents](https://luxor.ai/blog/hidden-ci-tax-ai-coding-agents.md): AI coding agents open more PRs and push more often, and every push runs CI. How to model the extra cost, measure the agent share, and cut it. - [Reviewing AI-generated code: a checklist for agent-written PRs](https://luxor.ai/blog/reviewing-ai-generated-code-checklist.md): Seven ways code from AI coding agents goes wrong while looking right, with an example of each and what to check before you approve the PR. - [Lint GitHub Actions workflows with actionlint: setup, config and CI](https://luxor.ai/blog/lint-github-actions-workflows-actionlint.md): Install actionlint, read its errors, set custom runner labels in .github/actionlint.yaml, and run it in CI with problem matchers, reviewdog or pre-commit. - [Monorepo CI with GitHub Actions: selective builds done right](https://luxor.ai/blog/monorepo-ci-github-actions-selective-builds.md): Run only the CI jobs a change affects, using path filters, Turborepo and Nx affected builds, dynamic matrices, required checks and runner sizing. - [Reusable workflows vs composite actions in GitHub Actions](https://luxor.ai/blog/reusable-workflows-vs-composite-actions.md): When to share GitHub Actions config as a reusable workflow or a composite action, covering secrets, runners, logs, nesting limits and versioning. - [GitHub Actions caching strategy: dependencies, builds and Docker](https://luxor.ai/blog/github-actions-cache-strategy.md): How the GitHub Actions cache matches keys, scopes and evicts entries, and how to cache npm, pnpm, Go, Python, Rust, Turborepo and Docker layers. ## Optional - [Runners benchmarks](https://luxor.ai/docs/runners/benchmarks.md): Benchmark results and methodology comparing Runners against GitHub-hosted runners across real open-source build workloads. - [Luxor Agent Cloud vs Blacksmith](https://luxor.ai/docs/runners/comparisons/tenki-vs-blacksmith.md): How Luxor Agent Cloud and Blacksmith compare on price, performance, free tier, macOS support, and migration. - [Luxor Agent Cloud vs Depot](https://luxor.ai/docs/runners/comparisons/tenki-vs-depot.md): How Luxor Agent Cloud and Depot compare on price, performance, free tier, macOS support, and migration. - [Brand Guidelines](https://luxor.ai/docs/brand-guidelines.md): Rules for using the Luxor Agent Cloud name, identity, and visual assets. - [Changelog](https://luxor.ai/docs/changelog.md): Product releases, improvements, and fixes. - [Luxor Agent Cloud Logos](https://luxor.ai/docs/branding.md): Downloadable Luxor Agent Cloud logos in supported formats. - [Referrals](https://luxor.ai/docs/built-with-tenki.md): Referral credit program and participation instructions. - [Socials](https://luxor.ai/docs/tenki-socials.md): Official community, support, and social channels. - [Luxor Agent Cloud pricing overview](https://luxor.ai/pricing.md): Pricing overview for Runners, Code Reviewer, and Sandbox. - [Luxor Agent Cloud for Startups](https://luxor.ai/startups.md): Program overview for eligible startup teams. - [AI Code Review Benchmark](https://luxor.ai/benchmarks/code-reviewer.md): AI code-review comparison and methodology. - [About](https://luxor.ai/company/about.md): Company background and team information. - [Careers](https://luxor.ai/company/careers.md): Current employment opportunities at Luxor Agent Cloud. - [Security](https://luxor.ai/company/security.md): Security and compliance overview. - [Complete documentation corpus](https://luxor.ai/llms-full.txt): Bulk-ingestion file containing all documentation; for normal tasks, prefer search or targeted Markdown pages because this file is large. - [Luxor Agent Cloud blog](https://luxor.ai/blog): Human-readable articles and product updates. - [Blog RSS feed](https://luxor.ai/blog/rss.xml): RSS feed containing all blog posts. - [Changelog Atom feed](https://luxor.ai/changelog.xml): Release feed with per-entry update timestamps. - [Luxor Agent Cloud on X](https://x.com/luxorcloud): Official Luxor Agent Cloud account on X. - [Luxor Agent Cloud on LinkedIn](https://www.linkedin.com/company/luxor-ai): Official Luxor Agent Cloud company page on LinkedIn.