Finally use AI on your sensitive files. Without the risk. Fully offline.

Import files with a simple drag and drop, or click to browse. Upload progress is tracked in real time with per-file status.

Works with PDF, DOCX, DOC, XLSX, PPTX, PNG, JPEG, TIFF, BMP, WebP, and more. Office files are automatically converted.

Select multiple files at once and process dozens of documents in a single batch.

Scanned PDFs and images are automatically processed with Tesseract OCR. word-level bounding boxes with intelligent line clustering.

Create folders inside the Library to group related files. All documents in the same folder are automatically linked and share one code registry.

Every file lives in a persistent Library you can revisit anytime. Search, sort, multi-select, favorite, and organize into folders that share one code registry. Separate Protect and Decode tabs keep your anonymization queue and AI-response files distinct, with a 30-day trash for safe deletes.

Patterns → Entities → Expressions → Deep Analysis, with automatic merging. Each layer catches what the others miss.

Finds names, organizations, locations, medical terms, and dozens of other context-dependent sensitive data types — the kind pattern matching alone can't catch. Choose from 14 specialized models, or let Auto mode pick the best one. Every model is scanned for malicious code before release.

Built-in patterns for SSNs, credit cards, IBANs, phones, business numbers, and more, tailored to 14 countries. Create your own with Simple block-builder or Advanced mode, plus a live test playground.

Run a local GGUF model on your GPU, or connect to any OpenAI-compatible API (OpenAI, Claude, Groq, Mistral, Ollama, vLLM) for deep context-aware analysis.

Enter specific words or phrases to find, paste from Excel or CSV. Each match can be set to Encode or Remove, with adjustable fuzziness.

Tune detection aggressiveness with a fuzziness slider, per-type toggles, and a region grouping control for how nearby words merge.

Country-specific patterns for the US, UK, Canada, Australia, France, Germany, Spain, Mexico, Italy, Netherlands, Belgium, Portugal, Brazil, and Switzerland. including business numbers, national IDs, and tax identifiers unique to each country.

Detected sensitive data automatically propagates across all pages, including partial org-name matching and type unification for consistency.

Pointer, lasso, and draw tools for selecting and creating regions. Merge, split, re-label, or delete with full undo/redo support.

Save and load named detection configurations (fuzziness, enabled types, expressions) to reuse across documents.

Choose per-region: encode (reversible replacement) or remove (permanent black-box redaction). Apply in bulk or one at a time.

Built-in noise filters reduce false positives for ORG, Location, Person, and Address types. Every detection includes a confidence score.

All document metadata, hidden fields, tracked changes, embedded objects, and invisible content are automatically stripped during processing. ensuring no sensitive data leaks through non-visible channels.

Review and protect documents without touching the mouse. Switch tools, act on regions, navigate pages, and undo — all from the keyboard.

Export encoded documents as clean PDFs: code text replaces sensitive data, removed regions are permanently redacted.

Export as PDF, DOCX, or XLSX. Office formats get in-place text replacement across paragraphs, tables, headers, and cells.

The Download popup exports all protected documents at once. Pick an AI tool preset (ChatGPT, Claude, Gemini…) and per-tool file count and size limits are set automatically. Up to 3 documents processed concurrently with real-time progress.

Multi-file exports are automatically packaged into a ZIP archive for easy download and sharing.

Split large exported PDFs into smaller parts for AI ingestion. Set a max size in MB. pages are intelligently grouped into parts.

Documents in the same folder automatically share a code registry: the same person, org, or address always gets the same code across every file in the folder. No manual linking required.

The Download popup lets you choose your AI tool — ChatGPT, Claude, Gemini, Copilot, Mistral, and more — and auto-sets the file count and size limits recommended for that service. An instruction prompt explaining every code prefix is embedded in the exported PDF and available to copy to your clipboard.

AI responses contain codes instead of real data, keeping sensitive information out of LLM logs and training sets.

Paste text or upload PDF, DOCX, XLSX, TXT, or CSV files, and promptShield replaces every code with its original value. The auto-generated AI prompt ensures LLMs preserve codes correctly, so decoding is seamless.

Codes are deterministic and consistent: the same entity always maps to the same code across the entire document set.

Review is mandatory: a file can't be downloaded until you've verified it, and only that review commits its codes to your persistent Code Registry. Nothing is ever saved silently.

Dashboard with total codes, documents, and registry size. Export the vault as JSON for backup, or import to restore.

After decoding, see exactly which codes couldn't be resolved, with count and full list for easy troubleshooting.

All processing and vault storage stays on your device. No sensitive data ever leaves your machine.

Process documents from your terminal with a single command. Detect, anonymize, and decode, all offline. Install with pip install promptshield-app.

Deploy the full REST API on your infrastructure with docker compose up. Interactive Swagger docs included. No data ever leaves your servers.

Import the core engine directly in your Python scripts. Fully async, type-hinted, and Pydantic-powered. Build custom workflows with full control.

Process hundreds of documents programmatically. Consistent code mappings across linked document sets. Integrate into CI/CD, ETL, or compliance workflows.

A live overview of every seat in your organisation: who's licensed, who has actually used promptShield recently, and where the gaps are.

Buy seats in bulk, see who occupies each one, and reassign licenses in a single click when someone leaves the company.

Every license event, role change, seat assignment, and significant action is recorded with actor, timestamp, and PII-free metadata.

Set the gap threshold (7, 30, 60 or 90 days) and the dashboard instantly reflags every member who has drifted past it.

One-click CSV export of the current adoption snapshot, ready to drop into an auditor's evidence pack or a board deck.

Pages processed, entities detected, and redactions applied — per member, per period — for capacity planning and ROI tracking.