# Documentation

- [Isorun](/): General-purpose execution runtime for AI agents. Sandboxes in milliseconds.
- Getting Started
  - [Quickstart](/getting-started/quickstart): Run code in a sandbox in three lines, then learn the primitives that make Isorun different.
  - [Authentication](/getting-started/authentication): API keys and how the SDK reaches the right region.
  - [Pricing](/getting-started/pricing): Per-second billing with runtime-first economics.
- Sandboxes
  - [Create & Execute](/sandboxes/create-and-execute): The default runtime loop for agents and jobs, create, execute, destroy.
  - [Lifecycle & Hibernation](/sandboxes/lifecycle): Create, run, hibernate, resume, and destroy a sandbox, and the cost model behind each transition.
  - [Persistent Storage](/sandboxes/persistent-storage): Attach a durable persistent disk to a sandbox.
  - [Checkpoints & Rollback](/sandboxes/checkpoints-and-rollback): Capture sandbox state, restore quickly, and fork prepared runtimes for parallel work.
  - [Shell Sessions](/sandboxes/detachable-sessions): Long-lived bash sessions inside a sandbox. State persists across commands.
  - [Port forwarding](/sandboxes/public-urls): Reach any port your in-sandbox code is listening on over HTTPS. Authenticated and WebSocket-aware.
  - [Any OCI Image](/sandboxes/any-oci-image): Use any Docker Hub / GHCR / private-registry image as a sandbox base. First boot pulls and builds; every subsequent boot is fast.
  - [Browser & computer use](/sandboxes/computer-use-vnc): Chromium with CDP and a live noVNC desktop inside a microVM. Drive it with Playwright or browser-use, watch it live, and run agent loops.
- Security
  - [Network Filtering](/security/network-filtering): Per-sandbox egress allow/deny lists. Block by default, allow what you trust.
  - [Network Profiles](/security/network-profiles): Named egress policy templates for common agent workloads.
  - [Endpoint Rules](/security/endpoint-rules): Restrict the host credential proxy to specific HTTP method + path combinations.
  - [Credential Injection](/security/credential-injection): Inject API keys into a sandbox without exposing the raw secret to the agent's environment.
  - [Audit Trail](/security/audit-trail): Tamper-evident NDJSON log of every event in a sandbox's lifetime. Per-tenant.
- SDKs
  - [TypeScript SDK](/sdks/typescript): Official Node.js client for sandbox lifecycle, command execution, file I/O, snapshots, fork, hibernate, and port exposure.
- Integrations
  - [Claude Managed Agents](/integrations/claude-managed-agents): Self-hosted sandbox runtime for Anthropic Claude Managed Agents. One microVM per session, isolated for the lifetime of the session.
  - [OpenAI Agents](/integrations/openai-agents): Run agent code inside Isorun sandboxes from the OpenAI Agents SDK.
  - [LangChain](/integrations/langchain): Use Isorun sandboxes as LangChain.js tools.
  - [CrewAI](/integrations/crewai): Run agent code inside Isorun sandboxes from CrewAI tasks.
  - [MCP Server](/integrations/mcp-server): Drive Isorun sandboxes from any MCP-compatible client (Claude Desktop, Cursor, Zed). Speaks JSON-RPC over stdio.

- **Reference**
- [API Reference](/api-reference): REST API for sandbox lifecycle, execution, files, snapshots, fork, hibernate, audit, and network policy.
- [Troubleshooting](/troubleshooting): Fixes for the errors and surprises you're most likely to hit.
- [Changelog](/changelog): What shipped and when.