# Atlas Labs > A free, anonymous, 30-minute API sandbox: two fictional REST systems (a Factory and a Head Office) for a printing company, Atlas Print Works & Co. Base URL https://automationatlas.io/labs/api/v1; send Authorization: Bearer ; keys start lab_fac_ (Factory) or lab_off_ (Head Office). Every number in every response is fictional simulation data (meta.simulation: true). Atlas Labs exists so automation tools and AI agents have something real to write to while testing, with no signup and no real-world consequences. It is curated by automation engineers who build these integrations for real businesses, is part of Automation Atlas, the independent site that reviews automation tools, is developed and funded by ShadowGen, and belongs to no vendor. ## Docs - [Atlas Labs home](https://automationatlas.io/labs/index.md): what the sandbox is, the two systems, and the three ways in. - [Quickstart](https://automationatlas.io/labs/docs/quickstart/index.md): four commands to your first authenticated write. - [Connect your tools](https://automationatlas.io/labs/docs/tools/index.md): the five fields every tool needs; which key goes where. - [Authenticate with your key](https://automationatlas.io/labs/docs/tools/rest-api/index.md): tested curl, JavaScript and Python for read, write, 429 and webhooks. - [Zapier](https://automationatlas.io/labs/docs/tools/zapier/index.md): Catch Raw Hook to receive, Custom Request to call, a Code step to verify signatures. - [n8n](https://automationatlas.io/labs/docs/tools/n8n/index.md): Webhook node (Raw Body) plus HTTP Request (Bearer Auth); public HTTPS on port 443 only. - [Make](https://automationatlas.io/labs/docs/tools/make/index.md): Custom webhook plus HTTP "Make a request"; API key keychain with the Bearer prefix. - [Power Automate](https://automationatlas.io/labs/docs/tools/power-automate/index.md): Swagger 2.0 custom connector or HTTP action; Request trigger set to Anyone. - [Moxo](https://automationatlas.io/labs/docs/tools/moxo/index.md): Moxo AI template Webhook trigger (None, secret URL) with field mapping and a condition; tested 2026-10-19. Moxo to Atlas Labs via a Make, n8n or Zapier hook. - [Frends](https://automationatlas.io/labs/docs/tools/frends/index.md): Frends.HTTP.Request task plus HTTP Trigger; not yet run in a tenant. - [AI agents (MCP)](https://automationatlas.io/labs/docs/tools/ai-agents/index.md): Streamable HTTP MCP endpoint; call create_sandbox, or send a key header; the Accept header gotcha. - [Waiting for a sandbox](https://automationatlas.io/labs/docs/queue/index.md): the 202 queue ticket, polling it, the claim window, 503 queue_full and Retry-After. - [Webhooks](https://automationatlas.io/labs/docs/webhooks/index.md): subscribe, verify the raw body, retries at 10 s, 60 s and 300 s. - [Concepts](https://automationatlas.io/labs/docs/concepts/index.md): the Factory/Head Office split, the bridge, autopilot, missions. ## About - [About Atlas Labs](https://automationatlas.io/labs/about/index.md): what it is, why it is free (a sandbox no vendor owns, curated by automation engineers), who runs it, exactly what is kept about a visitor and for how long, how to use it for a course or workshop (adjusted limits on request), and how to get in touch. - [Contact](https://automationatlas.io/labs/contact/index.md): feedback, bug reports, ideas, adjusted limits for a class and recipe suggestions, through a form a person reads; security reports go to cybersecurity@shadow-gen.com, with the form as the alternative. - [Security disclosure policy](https://automationatlas.io/labs/security/index.md): how to report a security problem (cybersecurity@shadow-gen.com), what is in and out of scope, the safe harbour, and what to expect back. ## Contact - [Contact](https://automationatlas.io/labs/contact/): feedback, bug reports, recipe suggestions and course set-ups. ## API reference - [OpenAPI 3.1](https://automationatlas.io/labs/api/v1/openapi.json) · [OpenAPI 3.0](https://automationatlas.io/labs/api/v1/openapi-3.0.json) · [Swagger 2.0](https://automationatlas.io/labs/api/v1/swagger-2.0.json) - [Self-describing index](https://automationatlas.io/labs/api/v1/): no key needed. ## Agents - [llms-full.txt](https://automationatlas.io/labs/llms-full.txt): the complete guide: create → status → act → back off, the common error codes, the tested snippets. - [MCP server](https://automationatlas.io/labs/mcp): Streamable HTTP; call create_sandbox, or send a key header. Read the labs://guide resource first (the agent loop), then call tools such as create_sandbox, sandbox_status and extend_sandbox. - [SKILL.md](https://automationatlas.io/labs/skills/atlas-labs/SKILL.md): the same loop as an Agent Skill (name and description front matter, then short instructions). ## Optional - [World status](https://automationatlas.io/labs/api/v1/world) · [History pack](https://automationatlas.io/labs/downloads/atlas-print-works-history.xlsx) · [Connector templates](https://automationatlas.io/labs/docs/tools/#downloads)