Privacy policy
This is a draft slappy document checked against the available implementation. It describes verified features and identifies unresolved matters. Production settings and actual operation require separate verification.
1. Operator and contacts
The operator of slappy is [OPERATOR_NAME], the service owner personally. Legal status: [OPERATOR_STATUS]. Country: [OPERATOR_COUNTRY]. Registration details: [OPERATOR_REGISTRATION]. Tax details: [OPERATOR_TAX_ID]. Postal address: [OPERATOR_POSTAL_ADDRESS]. These fields do not represent an already registered sole proprietor or company.
Website: [SITE_DOMAIN]. Data enquiries: [PRIVACY_EMAIL]. Support: [SUPPORT_EMAIL]. Until the placeholders are replaced, these are not working contact channels.
2. Data processed
Account data includes name, email, password hash, user ID, role, account status and email verification status. Email verification uses one-time codes and links; their hashes and expiry are stored in the database. A password is sent to the server when registering or signing in; the account record stores a hash.
Workspace materials include topics and instructions, AI prompts, scripts, uploaded HTML/JSON templates, images and music, voice settings, mode versions, projects, generated text, audio, subtitles and MP4 files. The data depends on the selected feature. Templates or prompts can include personal data you put in them.
Technical information includes job and project IDs, processing times and outcomes, errors, quota usage, a device identifier and information about IP addresses/login attempts for abuse prevention. Abuse records use hashed values; this does not make all logs anonymous. Network services also receive connection data.
Feedback includes a rating, willingness to pay, a comment and its association with a user and project. Enquiries involve the contact details and correspondence you provide. The payments/subscriptions collections in the code do not establish that purchases exist; actual billing status: [BILLING_STATUS].
If actual payments are introduced, the payment information received by the operator, what remains with the seller/provider, how receipts and webhooks are delivered, and record retention must be specified separately: [PAYMENT_DATA_FLOW]. This draft does not claim that the operator receives full card details, or that all payment data always stays exclusively with a provider.
3. Sources, purposes and legal bases
Data comes from you, operation of the service and selected media sources. It is processed to register and sign in, verify email, fulfil requests and generate content, store and display projects, apply quotas, protect the service, provide support and handle feedback. Required fields are indicated in forms; a feature may not work without necessary data.
Legal bases, operator/client roles and applicable law must be determined for each purpose: [LEGAL_BASES]. This draft does not treat consent as a universal basis or establish that GDPR applies to every visitor. Where processing relies on consent, separate collection and withdrawal mechanisms must be provided without affecting the lawfulness of prior processing.
Optional purposes, such as additional research, marketing or analytics, require separate descriptions of data, legal bases and choice mechanisms: [OPTIONAL_PROCESSING_PURPOSES]. They are not automatically necessary to fulfil an order. Decision on selling user materials/data: [CONTENT_SALE_POLICY]. Decision on training or evaluating the operator's own models: [OWN_MODEL_TRAINING_POLICY]. These matters are separate from external AI-provider terms; the draft asserts neither the presence nor absence of such processing and does not provide hidden permission for it.
4. Recipients and external services
When the relevant feature is used, the code calls OpenAI for scripts, structured content, HTML/JSON suggestions, web search and, in some modes, speech synthesis and transcription. Instructions, template text and structure, editing context, search queries, narration text or audio for transcription may be sent. Do not include information that is unnecessary for the task.
ElevenLabs receives text, the selected voice and synthesis settings when used. Resend receives the recipient address and verification email content, including a code or link. These are server integrations; availability depends on service configuration.
Pexels is used for stock video searches, Jamendo for music, and Iconify and its mirrors for icons. They receive search queries or identifiers of requested materials; downloads also contact the owners of the corresponding resources. OpenAI image searches may lead to downloads from other websites. This does not establish a licence for every result.
The main interface requests styles and fonts from Google Fonts: the browser connects to fonts.googleapis.com and fonts.gstatic.com, sending its IP address and technical request information. Locally bundled Roboto in certain template previews/renders does not remove these main-interface requests.
Authorised administrators and infrastructure providers may access data to operate and support the service. Provider account settings, contracts, roles, retention and data use require verification: [PROVIDER_SETTINGS]. We do not assert that every provider never uses data for training. Paddle is not integrated as a payment service or seller in the reviewed version.
Other disclosures and recipients, including binding lawful requests or business transfers, must be described for actual scenarios with applicable bases and safeguards: [OTHER_DISCLOSURES]. This is not unconditional permission to disclose all content to any third party.
5. Project and template visibility
The code checks ownership for personal projects and resources and also permits administrative access. This is not a promise that all information is accessible only to you. Do not put secrets or other people's confidential information in templates or prompts.
An administrator who owns a mode can publish a version in the shared catalogue. Other users can access the published template and resources included in that version; browser-accessible HTML previews can be inspected. Publishing a mode does not itself publish every personal project created with it.
Archiving a mode is reversible and retains data/versions. Existing projects and copied or exported materials do not automatically disappear when a mode is unpublished.
7. Location, retention and deletion
The code uses MongoDB, file storage for uploads and results, and Redis for jobs. Docker configuration does not establish the physical location of data. Application/database hosting: [HOSTING_PROVIDER], [HOSTING_LOCATION]. Files: [STORAGE_PROVIDER], [STORAGE_LOCATION]. Backups: [BACKUP_POLICY].
Retention periods or criteria: accounts — [RETENTION_ACCOUNT]; projects, uploads, versions and MP4 files — [RETENTION_CONTENT]; jobs, quotas, logs and abuse records — [RETENTION_LOGS]; feedback and enquiries — [RETENTION_SUPPORT]; payment records, if any — [RETENTION_BILLING]; backups — [RETENTION_BACKUPS]. Until completed, these fields do not establish specific retention periods.
Deleting an individual project, cleaning some files, expiry of a verification code and archiving a mode are different operations. The reviewed interface has no complete self-service procedure for deleting an account and all related data. An administrator deleting a user record does not establish erasure of all projects, logs, backups or provider-held data.
Verified deletion procedure and retention limitations: [DELETION_PROCESS]. Data requests should go to [PRIVACY_EMAIL] once that contact is configured. Instant, complete or guaranteed deletion within an arbitrary number of days is not promised; mandatory requirements of applicable law remain.
International transfers and applicable safeguards: [INTERNATIONAL_TRANSFERS]. An API integration or local database does not establish executed DPAs/SCCs, processing regions or third-party retention settings.
8. Data security
Verified technical and organisational safeguards, access controls, incident response and notification processes must reflect the infrastructure actually in use: [SECURITY_DESCRIPTION]. Password hashing, access checks or CSP in source code are not a promise of absolute security, end-to-end encryption or a particular certification.
9. Your rights and enquiries
Depending on applicable law, rights may include information and a copy of data, correction, deletion, restriction, objection, portability and withdrawal of consent. These rights have statutory conditions and exceptions. You may contact the competent supervisory authority; this document does not remove complaint or judicial remedies.
Request procedure, necessary identity checks and responses within applicable legal deadlines: [REQUEST_PROCESS]. Downloading an MP4 is not a complete personal data export. Settings allow changes to name and password, but do not provide every rights-request procedure.
The service uses automated quotas and abuse checks that may deny sign-in or generation. Questions and objections: [SUPPORT_EMAIL]. Processing for business clients and any required DPA/representative/DPO arrangements: [DPA_CONTACT]; this draft does not claim these are already in place.
10. Age and policy changes
Minimum age and rules for minors: [MINIMUM_AGE]; [CHILDREN_RULES]. They must match the markets served and actual service mechanisms; a registration form is not an age check.
Material processing changes must be reflected in a new policy version with accessible notice before the relevant processing where legally required. The dates and version in the header remain placeholders. Language rule: [LANGUAGE_RULE].