Privacy policy

How we measure usage

We use Umami, a self-hosted analytics service, to understand which tools are useful, how people find the site, and where requests fail.

Our analytics include:

• Visits: Page paths, referrers, browser type, and approximate country.
• Tool activity: Submission and result counts, selected options, exports, processing times, and error messages.
• Search events: Some tools, including Expiry and Reports, include entered search terms in their analytics events.

Analytics run on our infrastructure. Separately, our services and security providers process request metadata, including IP addresses, to operate the tools and limit abuse.

Third-Party Services

What runs under the hood

Browser Storage

What we store locally

We use browser storage for preferences, saved selections, cached results, and scan access. You can remove this data in your browser settings.

DNS Scout saves a random browser key and sends it with scan requests to control access to saved jobs and results. The server stores a hash of that key. Clearing that key removes this browser's access to those scans. When storage is blocked, access lasts only for the current page. Older scans created before this key was introduced cannot be reopened; start a new scan to get updated results.

Chat keeps its session, browser fingerprint, and conversation in tab storage so a reload can continue the chat. This browser copy is separate from the server records described below. Other security and session features may also use cookies.

Your Domain Data

What happens to your domains?

Each tool handles inputs differently. Domain lookups may send the domains you submit to DNS resolvers, registries, or other data providers. Our services may retain requests and results in caches, job storage, or operational logs.

Compare Domains: Your lists are compared in your browser. The comparison does not upload their contents to our servers.

File Splitter: Selecting a file uploads it to our server to read its columns and row count. Splitting uploads the file again for processing and returns the output for download. This tool does not offer saved file history.

DNS Scout: Saved deep scans include domains, results, a hashed browser identifier, and IP address. Completed and failed scan records become eligible for daily cleanup after 30 days; unfinished jobs may remain longer. Bulk lookup result files are separately eligible for cleanup after two hours. Queue records and logs can outlast the downloadable results.

Domain Appraiser: We cache domain appraisals for reuse for 30 days. Expiration stops reuse; physical removal happens during cache cleanup. Domains and processing details can also appear in service logs.

Secure Drop: Files are encrypted on our server for operator access. By default, files expire after 72 hours and cleanup runs hourly. The upload page shows the current retention and cleanup interval.

The AI assistant (Gizmo) works differently; see below.

Calls to api.toyroom.ai

The Toyroom API and the Toyroom connector for Claude and other AI assistants run at api.toyroom.ai. They receive the domain names and keywords in each request. They never receive your conversation, your files, or your assistant's memory.

• •What we receive: the tool or endpoint called and its inputs, such as domain names, keywords, and extensions. In an AI assistant, the assistant decides what to send. Requests from Claude on the web, desktop, and mobile come from Anthropic's servers.
• •How we use it: to answer the request, enforce rate limits and daily allowances, and keep the service running. We don't sell it or use it for advertising.
• •Rate limits: the keyless connector counts requests per caller: one IPv4 address, or one IPv6 /64 (and its /48). We store a keyed hash of the address that changes every day, never the address itself. Live counters expire after 26 hours. API keys are stored as hashes, and daily usage per key expires after 26 hours. Nightly backups of our database keep copies of these records for up to 60 days.
• •Logs: web server logs record each request's IP address, time, request line, status, referrer, and browser, not its body. Service logs record each request's path and status, and errors, which can include a short part of an input we rejected. Both rotate at a fixed size.
• •Bulk jobs: lists submitted with an API key, and their results, are deleted 24 hours after the job finishes.
• •Other services: Cloudflare carries all traffic to api.toyroom.ai. RDAP and DNS lookups, available with an API key, send domain names to registries and DNS resolvers. The keyless connector answers from Toyroom's own model and zone index.

Backups expire on their own schedule, within 60 days. For privacy questions or to ask us to delete records, write to [email protected].

AI Assistant

How Gizmo handles your data

Gizmo, our AI assistant, uses OpenAI's API to help you with domain-related questions. It stores conversation history and session information to maintain context and enforce usage limits.

• •Temporary storage: Active conversation and session records expire after 24 hours without a refresh. Chat activity can extend this period. Session records include your IP address, browser fingerprint, and usage counts.
• •Service logs: Audit logs can include session identifiers, excerpts of messages, tool arguments, and processing details. Logs and backups may retain records longer than active chat storage.
• •OpenAI processing: Messages and tool results are sent to OpenAI to generate responses. Starting a new chat does not remove existing records from service logs or backups.

For details on how OpenAI handles API data, see their privacy policy.