Privacy Policy
Yunyu Chengye is provided by Cixi Yunyulai Culture & Technology Co., Ltd. YunyuTrace provides file registration and verification capabilities.
Files stay on your device
Image decoding, screenshot pagination, PDF generation, watermark processing and digest calculation take place locally in your browser. Original images, PDFs, DOCX files, document bodies, chat content and OCR content are not sent to this service. Files and previews from the current task are not retained as cloud history after you close the page.
Information we retain
To support verification, the service stores final file digests, content fingerprints, hashed verification codes, file identity and signature node IDs, page counts, file sizes, generation types, registration times and verification counts. Signing after compression also records the Chengye processing relationship between the original and compressed registrations. Full layout manifests, original images and original filenames are not stored in the database. For automatic verification, external signed evidence packages, page dimensions, salted page commitments, page order and signed page roots are stored in COS. These objects are encrypted on the server and contain no page images. Proof archives use a fixed product filename; the database stores hashed verification codes.
Payments and anonymous analytics
Paid exports record the order number, actual order amount, payment status, transaction ID, hashed export authorization and necessary timestamps. Payments are processed by the payment provider. Anonymous analytics record feature events and quantity ranges only. They do not include file content, file names, verification codes or file digests linked to user behavior.
Verification and data retention
Anyone holding a verification code can look up its registration records and compare local file digests to check for changes. Proof is returned only when both the verification code and a matching file digest are provided. The browser automatically compares rendered pages locally. Registration records support future verification; order records support payment reconciliation. You are responsible for saving original and generated files. The service does not offer file recovery or cloud storage.
No account required
No registration, login or phone number is required to use the tools. Each operation is a separate task. Save your files and verification codes before leaving the page.
Language preference
A first-party cookie remembers the interface language for payment return pages. It contains only the language code and expires after one year. It is not used for advertising or cross-site tracking.
PDF redaction processing
PDF redaction accepts PDF files only. A local browser worker extracts readable text and its page positions, identifies candidate private information and previews text selections or manually marked areas with solid or rainbow masks. Source files, original filenames, extracted text, page images, positions, custom terms, selections and edit history are not sent to the service or stored in a cloud history. The browser renders each page locally with its masks applied and creates a high-resolution image-only PDF with the original page dimensions. The export omits the original text layer, source metadata, annotations, attachments, signatures, old file identity and unused source content. The local preview downloads the PDF directly and displays a simulated verification result. The online service adds a new YunyuTrace identity to the final PDF. Registration sends final file digests, content fingerprint, file size, page count and the new identity and signature node identifiers. Necessary verification proofs are encrypted and saved; proof archives use a fixed product filename. Registration excludes original body text, custom private terms and the original filename. Unsupported content that cannot be reliably processed causes export to stop. Cancelling, changing the source file or leaving the page ends the worker session. There is no automatic fifteen-minute idle expiry; task timeouts apply only to running operations.
YunyuTrace processing declarations in output files
Some newly generated or signed PDFs store a YYT-PROCESSING-STATEMENT/1 declaration in the non-page catalog to describe the operation, local processing, input count and output pages to file holders. Redaction adds the rule version, candidate and selection counts, manual regions, pages without text, mask appearance and absence of OCR. The declaration excludes source text, matched values, terms, positions, per-page lists, category details, source filenames and digests, user identities and credentials, and adds no pages or attachments. Counts can reveal processing activity and accompany the downloaded file; consider recipients before sharing. The declaration is not sent as new separate registration fields. Final digests bind the output bytes containing it; signatures and page proof do not independently establish declaration truth, complete redaction, content legality or processing authority.
Deployment and network access
The production application and registration support services are deployed in Tencent Cloud Shanghai Zone 2. Verification evidence is archived through the Tencent Cloud COS Shanghai regional endpoint. This describes the production deployment and archive endpoint, not the location of every network edge or infrastructure replica. Website access and registration involve network information such as IP addresses, request times and necessary identifiers for communication, abnormal-request limits and service security, not PDF detection or user profiling.
Runtime diagnostics
The service collects API outcomes, runtime failures and categories of tool errors and notices for troubleshooting. Logs include time, severity, page or API category, error codes, sanitized source locations, duration and random request identifiers. They exclude source files, document content, original filenames, verification codes, access credentials, keys, complete request parameters and custom privacy terms. Logs are archived through the existing Tencent Cloud COS storage and retained for 30 days, with expired records removed by the running cleanup task. When storage is unavailable, a bounded server queue holds logs for later upload; expired queued logs are not uploaded. Logs are not used for advertising, profiling or model training.
Scope of registration and proof data
For PDF redaction, the content fingerprint is computed from the output after selected content is removed, and final file digests from the file after the new YunyuTrace identity is added. Random salts, dimensions, order and commitments in the full page proof are archived, not exclusively kept in the browser. Archive writing requires a chain access credential. Retrieval requires a matching verification code, file digests and size; a code alone does not retrieve the full proof. Digests and registration metadata may relate to a particular file and are not necessarily anonymous.
Privacy rights and retention
Use the complaints and privacy requests channel to ask about information held by the service, request access, a copy, correction, supplementation or deletion of related personal information, withdraw relevant consent or request restrictions. We review the relationship, authority and necessary retention basis for the specific matter and explain why a request cannot be fulfilled. Registration and proof records support continuing verification; there is currently no fixed-day automatic deletion mechanism for these records. Source files, previews and local editing history are not stored on the server and cannot be recovered or deleted from your device by us.
Complaints and privacy contact
Provider: 慈溪云屿来文化科技有限公司。Responsible person: 赖海波。Telephone: 18267400651;Email: laihaibo@yunyulai.cn。
Start with the date, issue type and page or region reference. Provide the processed file’s SHA-256 digest only if a registered version must be located. Do not send the source PDF, complete sensitive numbers, verification codes, passwords or access credentials. Complaint messages, reply details and necessary review records are used for handling and communication, not advertising, profiling or algorithm training. If an infrastructure provider must assist, unrelated content is removed and the scope is limited.
We normally acknowledge a contactable, understandable request within 3 working days and provide a result within 15 working days of acceptance. For unresolved complex issues, we explain the reason, progress and next steps. This procedure applies from publication of the feedback page. See Complaints and privacy requests。