Noulette is a MCP platform that finds problems, propose fix and actually fix it (upon approvals) during app development and when running app in production.
FAQ (answers come from our records)
How do I install it?
We did not find an install command for this repository (our install coverage is only about 4%). Check its README or releases page.
Who is it for?
We filed it under these tags: 开源项目. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository vibecoder0715/Noulette;first catalogued 2026-10-10;last seen 2026-10-10。There is no official certification or quality rating here — we record sources and dates only.
Noulette
以下摘录自该仓库的 README(片段) · 在 GitHub 读完整 README →
Finds problems in your app before it ships, and keeps watching it after launch.
Pitch deck: **Pitch deck**
1. Download
**Download Noulette for Windows (latest .exe)** · Website
*(Always the newest release from GitHub Releases | v1.0.4 SHA-256: 0504efc54f07d6ba57c4163a899b5609152b8fec00669becf91bf5df8c7e3c0e)*
One installer for everything: the desktop app, the MCP server your coding agent talks to, and the bundled scanners. Windows 10/11 x64, installs for the current user, no admin rights or extra runtimes needed.
The installer is not code-signed yet, so Windows SmartScreen will show "Windows protected your PC". Click **More info**, then **Run anyway**. Only do this for a file you downloaded from the official link.
2. Mission and vision
Noulette does not just check, detect problems and propose decisions about your code before pushing to production. When enabled with an API key, it actively checks logs, inspects and detects or predicts log issues, and proposes fixes directly in your notifications for approval. Nothing changes without your permission.
3. Core functions
- **Inspect**: Scans your code for security leaks, vulnerable packages, and logic flaws before launch.
- **BUILD**: Fixes approved issues, re-runs tests, and verifies the solution automatically.
- **Auto-Approve**: Optionally inspects and fixes issues on an isolated branch.
- **Download Guard**: Quarantines suspicious files and downloads to keep your project safe.
- **Active Log Monitoring**: Watches errors and traffic, predicts problems, and proposes fixes for approval.
- **Future Plans**: Maps out scaling requirements as your user base grows.
4. Environments
- **Coding Agents**: Connects with Antigravity, Claude Code, Cline, Cursor, and any MCP tool.
- **Analytics**: Supports PostHog and Google Analytics 4.
- **AI Models**: Works with OpenAI-compatible APIs and local models like Ollama.
- **Local & Private**: Runs locally on your PC. Your data and keys stay secure.
- **Cloud Worker**: Optional self-hosted worker to monitor logs when your PC is offline.
- **Build from Source**: Runs on Bun and Rust (bun install, bun test, bun run build).
5. Sample training script
Sample training script for the agent that **monitor, detect and predict the logs** are as in (trainings folder), details described in (trainings/README.md).