Releases: AmaLS367/FaceScore
FaceScore 0.1.1
🚀 FaceScore v0.1.1
FaceScore v0.1.1 is a security-focused update for the desktop app.
This release keeps the same clean AI-powered appearance analysis workflow, but fixes the API key storage model: your Anthropic API key is now saved through the operating system credential store and reused after restarting the app.
No repeated key entry. Less JavaScript exposure. A cleaner desktop security boundary.
✨ Highlights
- 🔐 Secure API key storage via OS credential store
- 🔁 API key persists after app restart
- 🦀 Anthropic requests moved into Rust commands
- 🧼 API key no longer travels through the normal analysis flow in JavaScript
- 🛡️ Removed direct WebView HTTP access to Anthropic
- ✅ Updated tests for the new secure key flow
📦 What’s included
This release includes the secure key-storage update on top of the v0.1.0 public build:
- OS-backed API key persistence
- startup check for whether a key is already configured
- secure save and clear actions for the Anthropic API key
- Rust-side Anthropic Messages API request handling
- frontend analysis flow without passing the stored key around
- updated CSP and Tauri capabilities
- updated Windows installers for v0.1.1
⬇️ Downloads
Choose the file that fits your setup:
.exe→ standard Windows installer.msi→ Windows package installer- update metadata / extra assets → internal release files for app updates and distribution
If you just want to install and use FaceScore normally, download the .exe build.
⚡ Quick start
- Download the installer from this release
- Install and launch FaceScore
- Paste your Anthropic API key once
- Upload a photo
- Click Analyze face
- Review the generated report
- Restart the app and continue without re-entering the key
- Export the report to PDF if needed
🔐 Security update
Previous builds kept the API key only in the JavaScript runtime for the current session. That avoided disk persistence, but it also meant users had to re-enter the key after every restart and the key moved through the WebView analysis path.
In v0.1.1:
- the key is stored using the OS credential store
- the app checks only whether a key exists on startup
- the Anthropic request is performed from Rust
- the stored key is not passed into the frontend analysis function
- the Tauri HTTP plugin and direct Anthropic WebView capability were removed
The key can still briefly exist in the frontend while the user types it into the password field. After saving, the input is cleared.
💎 What FaceScore is good for
FaceScore works especially well if you want to:
- improve your profile photo
- get more structured appearance feedback
- build a self-improvement baseline
- compare photos over time
- keep feedback private on your own machine
- avoid re-entering your API key every session
🛡️ Notes
- Requires your own Anthropic API key
- API key is stored through the operating system credential store
- Supports JPG, PNG, and WebP
- Maximum image size: 5 MB
- Reports are generated through Anthropic's API from the desktop app
- No project-operated backend is used for API keys or photos
🔥 Why this release matters
This release closes the biggest practical security and UX gap in the first public build.
The app now behaves more like a proper desktop tool: configure your key once, keep it in the OS credential store, and run analysis without exposing the stored key through the normal JavaScript request path.
🧪 Verification
The v0.1.1 build was checked with:
- TypeScript typecheck
- Vitest test suite
- ESLint
- production frontend build
- Rust unit tests
- Tauri Windows bundle build
💬 Feedback
Tried FaceScore?
- 🐞 Found a bug: open an issue
- 💡 Have an idea: request a feature
- ⭐ Like the project: star the repo
Your feedback helps shape the next release.
FaceScore 0.1.0
🚀 FaceScore v0.1.0
The first public release of FaceScore is here.
FaceScore is a privacy-first desktop app for AI-powered appearance analysis.
Upload a photo, get a structured report, review strong and weak points, and export everything to PDF.
No messy browser tool vibe. No random opinions. Just a clean desktop workflow with practical feedback.
✨ Highlights
- 🤖 AI-powered face analysis
- 📊 Structured appearance report
- 📄 PDF export
- 🛡️ Local-first privacy approach
- ✅ Strict response validation
- 🖥️ Desktop app experience
📦 What’s included
This release includes the first usable public build of FaceScore with:
- photo upload
- Claude Vision analysis
- structured report rendering
- score and feedback sections
- PDF export flow
- local API key usage
- schema validation with Zod
⬇️ Downloads
Choose the file that fits your setup:
.exe→ standard Windows installer.msi→ Windows package installer- update metadata / extra assets → internal release files for app updates and distribution
If you just want to install and use FaceScore normally, download the .exe build.
⚡ Quick start
- Download the installer from this release
- Install and launch FaceScore
- Paste your Anthropic API key
- Upload a photo
- Click Analyze face
- Review the generated report
- Export it to PDF if needed
💎 What FaceScore is good for
FaceScore works especially well if you want to:
- improve your profile photo
- get more structured appearance feedback
- build a self-improvement baseline
- compare photos over time
- keep feedback private on your own machine
🛡️ Notes
- Requires your own Anthropic API key
- Supports JPG, PNG, and WebP
- Maximum image size: 5 MB
- Reports are generated through Anthropic's API from the desktop app
- API keys are not meant to live in a project-operated backend
🔥 Why this release matters
This is the first public version of FaceScore and the foundation for future updates.
The goal is simple:
make appearance feedback feel structured, private, and actually useful.
🧪 Early release status
This is an early public release, so expect future improvements in areas like:
- richer report presentation
- better photo-specific guidance
- comparison features
- more export and customization options
💬 Feedback
Tried FaceScore?
- 🐞 Found a bug: open an issue
- 💡 Have an idea: request a feature
- ⭐ Like the project: star the repo
Your feedback helps shape the next release.