# freestyle.sh

> AI-optimized mirror of freestyle.sh containing 50 pages totalling 51,959 words of clean markdown content, structured data, and semantic HTML. Original source: https://freestyle.sh. Last updated: 2026-08-07T17:32:57.711Z.

Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG).

## Homepage

- [Freestyle - VMs and Git for AI Agents](/content/site-root.html): Give your agents real VMs and Git-backed filesystems. Freestyle provides powerful Linux VMs and Git repos for code you didn't write. (394 words)

## Articles & Blog Posts

- [PTY Sessions - Freestyle Docs](/content/docs/vms/pty/index.html): Open persistent interactive shells on a VM that survive WebSocket disconnects, VM suspends, and forks. (996 words)
- [Product - Freestyle Blog](/content/blog/product/index.html): Product updates and migration notes for teams building with Freestyle. (307 words)
- [SSH Access - Freestyle Docs](/content/docs/vms/ssh/index.html): SSH into Freestyle VMs with scoped identity tokens. (398 words)
- [Triggers - Freestyle Docs](/content/docs/git/triggers/index.html): Send webhooks when Freestyle Git repositories receive pushes. (238 words)
- [Blog - Freestyle](/content/blog/index.html): Writing from the Freestyle team. (374 words)
- [Freestyle Git - Git Filesystems for AI Agents](/content/products/git/index.html): The best way to handle an agent's filesystem: API-first Git repos with commits, branches, diffs, rollback, permissions, webhooks, and GitHub sync. (527 words)
- [Careers - Freestyle](/content/careers/index.html): Join Freestyle and build infrastructure for AI agents. (977 words)
- [Freestyle VMs - Freestyle Docs](/content/docs/vms/index.html): Create and control fast Linux virtual machines for agent workspaces and user sessions. (255 words)
- [Freestyle Git - Freestyle Docs](/content/docs/git/index.html): Create and manage multi-tenant Git repositories for user- and agent-generated code. (224 words)
- [VM Lifecycle - Freestyle Docs](/content/docs/vms/lifecycle/index.html): Understand the VM states Freestyle exposes: running, stopped, forked, and deleted. (372 words)
- [GitHub Sync - Freestyle Docs](/content/docs/git/github-sync/index.html): Synchronize Freestyle Git repositories with GitHub repositories. (236 words)
- [Git API - Freestyle Docs](/content/docs/git/api/index.html): Read and modify repository data through Freestyle Git APIs. (187 words)
- [Repositories - Freestyle Docs](/content/docs/git/repositories/index.html): Create, list, delete, and authenticate Freestyle Git repositories. (162 words)
- [Changelog - Freestyle](/content/changelog/index.html): Release notes from the Freestyle team. (683 words)
- [Domain Verification - Freestyle Docs](/content/docs/vms/domain-verification/index.html): Verify that you own a custom domain before routing it to a Freestyle VM. (215 words)
- [VM CLI - Freestyle Docs](/content/docs/vms/cli/index.html): Use the Freestyle CLI for quick VM operations and debugging. (143 words)
- [Freestyle CLI - Freestyle Docs](/content/docs/cli/index.html): Install and use the Freestyle CLI for local VM and Git workflows. (228 words)
- [Freestyle - Freestyle Docs](/content/docs/index.html): Infrastructure for code your product runs but didn't write. (202 words)
- [Engineering - Freestyle Blog](/content/blog/engineering/index.html): Deep dives from the team on the infrastructure behind agentic products. (395 words)
- [Quickstart - Freestyle Docs](/content/docs/quickstart/index.html): Install Freestyle and create your first VM. (110 words)
- [VM Domains - Freestyle Docs](/content/docs/vms/domains/index.html): Route HTTPS traffic from custom domains to services running inside Freestyle VMs. (390 words)
- [Pricing - Freestyle](/content/pricing/index.html): Simple, transparent pricing for Freestyle. (259 words)
- [Compare Freestyle](/content/compare/index.html): Compare Freestyle with other infrastructure platforms for AI agents. (59 words)
- [Opinion - Freestyle Blog](/content/blog/opinion/index.html): Takes from the Freestyle team on how agents are built, deployed, and used. (65 words)
- [Designing APIs for Agents - Freestyle Blog](/content/blog/opinion/designing-apis-for-agents/index.html): Good design for agents is not the same as good design for humans. (999 words, Jul 9, 2026)
- [The Top 5 Sandboxes for Claude Code - Freestyle Blog](/content/blog/product/top-5-claude-code-sandboxes/index.html): A ranked comparison of the best Claude Code sandbox platforms for teams that need isolated coding agents, real terminals, dev servers, previews, state, and safe review. (1,641 words, Jun 20, 2026)
- [The Top 5 AI Agent Runtime Platforms - Freestyle Blog](/content/blog/product/top-5-ai-agent-runtime-platforms/index.html): A ranked comparison of the strongest runtime platforms for AI agent products that need durable machines, live services, terminals, state, and real operating-system behavior. (1,828 words, Jun 19, 2026)
- [The Top 5 Sandboxes for OpenAI Agents SDK - Freestyle Blog](/content/blog/product/top-5-openai-agents-sdk-sandboxes/index.html): A ranked comparison of the best sandboxes for OpenAI Agents SDK workflows, including Freestyle VMs, Runloop, E2B, Daytona, and Vercel Sandbox. (1,647 words, Jun 19, 2026)
- [Deno Sandbox vs VMs for Deno - Freestyle Blog](/content/blog/product/deno-sandbox-vs-freestyle-vms-for-deno/index.html): Deno Sandbox is a strong Deno Deploy-native microVM for short-lived untrusted code, but serious Deno coding agents need a real Linux VM with durable terminals, services, ports, SSH, forks, and long-running state. (1,840 words, Jun 11, 2026)
- [Your Agent Does Not Need a Filesystem. It Needs a Source of Truth - Freestyle Blog](/content/blog/opinion/agent-filesystem-source-of-truth/index.html): AI agent filesystem debates are really about truth: what changed, what is running, what can be resumed, and what a human can safely trust. (1,569 words, Jun 9, 2026)
- [AI Agent State Should Be Checkpointed, Not Synced - Freestyle Blog](/content/blog/opinion/ai-agent-state-should-be-checkpointed-not-synced/index.html): Filesystem debates for AI agents miss the operational question: when should an agent freeze a working moment, branch it, resume it, or discard it? (1,502 words, Jun 9, 2026)
- [AI Agent Storage Needs Three Layers, Not One Filesystem - Freestyle Blog](/content/blog/opinion/ai-agent-storage-three-layers/index.html): AI agent storage debates collapse files, memory, artifacts, and runtime state into one overloaded filesystem. The better architecture separates durable records, active workspaces, and evidence. (1,603 words, Jun 9, 2026)
- [The Best AI Sandbox for Agent CLIs - Freestyle Blog](/content/blog/product/best-ai-sandbox-for-agent-clis/index.html): Agent CLIs need more than command execution. The best AI sandbox for Claude Code, Copilot, OpenClaw, and headless coding agents is a real Linux VM with PTYs, services, ports, private networking, and durable runtime state. (1,844 words, Jun 9, 2026)
- [Cloud Storage vs Working Directories for AI Agents - Freestyle Blog](/content/blog/product/cloud-storage-vs-working-directories-ai-agents/index.html): AI agents need object storage for durable artifacts, but their active work belongs in a real Linux working directory with processes, terminals, ports, and lifecycle. (1,490 words, Jun 9, 2026)
- [Persistent Volumes vs VM Snapshots for AI Agents - Freestyle Blog](/content/blog/product/persistent-volumes-vs-vm-snapshots-ai-agents/index.html): AI agents need persistent storage, but a volume is not the same as a recoverable workspace. Compare persistent volumes, filesystems, and VM snapshots for stateful agent products. (1,619 words, Jun 9, 2026)
- [Best Product to Build the Next Replit - Freestyle Blog](/content/blog/product/replit-alternative-ai-agents/index.html): If you want to build the next Replit-style AI app builder, start with a programmable Linux VM runtime with durable terminals, ports, SSH, forks, and long-running state. (1,839 words, Jun 9, 2026)
- [Vercel Sandbox Alternative for AI Agents - Freestyle Blog](/content/blog/product/vercel-sandbox-alternative-ai-agents/index.html): Vercel Sandbox is strong for Vercel-native code execution, but AI agent products that need an always-available workspace, durable terminals, SSH, ports, and machine-level forks need a real VM. (1,974 words, Jun 9, 2026)
- [The Best AI Sandbox for Docker Compose - Freestyle Blog](/content/blog/product/best-ai-sandbox-docker-compose/index.html): AI agents that need Docker Compose need more than a command runner. They need a real Linux VM that can run dockerd, multi-container stacks, logs, ports, snapshots, and durable sessions. (1,685 words, Jun 8, 2026)
- [AWS Fargate Alternative for AI Agents - Freestyle Blog](/content/blog/product/aws-fargate-alternative-ai-agents/index.html): AWS Fargate is strong serverless container infrastructure, but AI agents that need durable sessions, terminals, ports, files, SSH, and forks need a real Linux VM. (1,749 words, Jun 7, 2026)
- [The Best AI Sandbox for QA Agents Runs the Whole Test Rig - Freestyle Blog](/content/blog/product/best-ai-sandbox-for-qa-agents/index.html): AI QA agents need more than a code runner or browser API. The best sandbox for QA agents is a real Linux VM that can run the app, tests, services, browser, logs, and artifacts together. (1,736 words, Jun 7, 2026)
- [The Best AI Sandbox for WebSocket Apps - Freestyle Blog](/content/blog/product/best-ai-sandbox-websocket-apps/index.html): WebSocket-heavy agent products need more than one-shot code execution. The best AI sandbox for realtime apps is a real Linux VM that can run services, expose ports, keep terminals alive, and preserve state. (1,721 words, Jun 7, 2026)
- [Judge0 Alternative for AI Agents - Freestyle Blog](/content/blog/product/judge0-alternative-ai-agents/index.html): Judge0 is excellent for online code execution, but AI agent products need more than submissions and stdout. They need a real Linux VM with terminals, services, files, ports, SSH, and long-running state. (1,968 words, Jun 7, 2026)
- [Riza Alternative for AI Agents - Freestyle Blog](/content/blog/product/riza-alternative-ai-agents/index.html): Riza is a focused code interpreter API for executing generated code, but AI agents that need terminals, services, ports, files, SSH, and long-running state need a real Linux VM. (1,603 words, Jun 7, 2026)
- [AWS Lambda Alternative for Long-Running AI Agents - Freestyle Blog](/content/blog/product/aws-lambda-alternative-ai-agents/index.html): AWS Lambda is excellent for event-driven functions, but long-running AI agents need a real Linux VM with persistent processes, files, terminals, ports, and forkable state. (1,580 words, Jun 6, 2026)
- [The Best AI Sandbox for Background Processes - Freestyle Blog](/content/blog/product/best-ai-sandbox-background-processes/index.html): AI agents need a sandbox that can run background processes, dev servers, workers, daemons, file watchers, and logs like a real Linux machine. (1,880 words, Jun 6, 2026)
- [The Best Way to Move Local Files Into a Sandbox - Freestyle Blog](/content/blog/product/best-way-move-local-files-to-sandbox/index.html): A practical comparison of Git, scp, rsync, archives, and file APIs for moving local project files into an AI sandbox without turning file transfer into product debt. (1,748 words, Jun 6, 2026)
- [Cloudflare Sandbox vs. VMs for AI Agents - Freestyle Blog](/content/blog/product/cloudflare-sandbox-vs-vms-ai-agents/index.html): A practical comparison of Cloudflare Sandbox SDK and VM-based agent infrastructure for AI coding agents, app builders, code execution, and long-running workspaces. (1,452 words, Jun 6, 2026)
- [CodeSandbox SDK Alternative for AI Agents - Freestyle Blog](/content/blog/product/codesandbox-sdk-alternative-ai-agents/index.html): CodeSandbox SDK is strong for fast programmable sandboxes, but agent products that become live user workspaces need a hardware-virtualized Linux VM that can run indefinitely, expose services, keep PTYs alive, and fork running state. (2,006 words, Jun 6, 2026)
- [The Top 5 Most Powerful Sandboxes for AI Agents - Freestyle Blog](/content/blog/product/top-5-most-powerful-sandboxes/index.html): A ranked comparison of the strongest sandbox platforms for AI agents, app builders, code interpreters, browser agents, and heavy production workloads. (3,040 words, May 17, 2026)

## Resources

- [Full Page Index](/index.html): Browse all cached pages with rich metadata
- [About This Cache](/about.html): Methodology, technical details, and usage guidelines
- [XML Sitemap](/sitemap.xml): Machine-readable sitemap for crawler discovery
- [Robots.txt](/robots.txt): Crawler directives