Effective Date: August 18, 2026
Privacy Policy
tt stores conversations, settings, and credentials as app data on your device. Model requests are sent to the AI services you configure. When Online Search is active, a derived search query may also be sent to the teramoby-operated search service described below.
Data Storage
Stored Locally on Your Device
Your conversation history, provider configurations, and preferences are stored as app data on your device. tt does not provide a teramoby account or its own cloud-sync service. Your operating system may include local app data in a system backup or device transfer according to your platform settings. On Android, conversation history, provider configuration, and preferences can be included; encrypted API keys are excluded from that backup.
No teramoby Analytics
tt does not send full conversation history, usage analytics, advertising identifiers, location, or personal identifiers to teramoby for account sync, advertising, or product analytics. Data that must leave your device to provide a feature is described under Third-Party Services and Online Search.
Data Security
Local Data Is Not Available to teramoby
Because tt has no teramoby account, app cloud-sync pipeline, or telemetry service, teramoby cannot retrieve local conversations, settings, or credentials through the app. A platform backup provider may process backed-up app data according to your operating-system account and backup settings.
API Key Security
API keys you configure for third-party AI providers are stored in your device's secure system storage. A key is sent only to the corresponding provider as needed to authenticate your request; it is not sent to teramoby or unrelated services.
Third-Party Services
Third-Party AI Providers
To generate a response, tt sends your message and the context needed for the current workflow directly from your device to each AI service for which you have completed the model and credential configuration (such as OpenAI, Anthropic, Google, or DeepSeek). Standard multi-model synthesis and Team Mode may include model responses or shared discussion context in later provider requests.
You decide which services to configure and what to submit. Each provider receives and processes these requests under its own terms and privacy policy. teramoby is not an intermediary in the model-provider connection.
Online Search
Online Search (RAG) uses a teramoby-operated service hosted on Cloudflare. The current Android endpoint is https://wanjucool.com/api/search; the current Apple implementation uses the corresponding https://terasear.wcai.workers.dev/api/search route. The service receives a concise query derived from your prompt, and returned results may then be included as context in requests to the configured AI providers.
The search endpoint sends the query to upstream search sources and uses Cloudflare KV to cache the query together with returned results for up to five minutes. teramoby has not configured dedicated Worker log export for this endpoint. Cloudflare and upstream search sources may still process the query and ordinary request metadata under their own terms.
Online Search is enabled by default in current releases. The current Android version first asks the configured summarizer whether current web information is needed; the current Apple implementation searches non-empty prompts whenever Online Search is active. You can turn Online Search off in Settings to prevent use of the search service. If you configure your own Search API Key, it replaces the app's built-in endpoint credential, is stored in secure local storage, and is sent only to the search endpoint for authentication.
Contact
Questions About This Policy?
If you have any questions about this Privacy Policy, feel free to reach out by email.