- Fix timingSafeCompare to pad buffers to equal length before comparing,
closing the content timing side-channel on length mismatch
- Replace innerHTML with textContent for error messages to prevent XSS
- Extract inline script from dashboard/index.html into dashboard/app.js
- Remove 'unsafe-inline' from script-src CSP directive
- Update security test to match new CSP policy
https://claude.ai/code/session_01GfwUNbgeMgXreFDbsNQMsS
Adds standard HTTP security headers (CSP, X-Content-Type-Options, X-Frame-Options, Referrer-Policy) to all dashboard responses to mitigate XSS, clickjacking, and MIME sniffing attacks.
Includes verification tests in test/dashboard-security.test.mjs.
Co-authored-by: PR0M3TH3AN <[email protected]>
Replaces synchronous `fs.readFileSync` with `fs.promises.readFile` in `loadTorchConfig` to prevent event loop blocking.
Updates all dependent configuration getters and consumers to use async/await.
Includes benchmark proving non-blocking behavior.
Fixes dashboard tests to properly await server listening.
Co-authored-by: PR0M3TH3AN <[email protected]>
This patch addresses a security vulnerability where the dashboard server could serve any file from the package root.
Changes:
- Implemented an allowlist in `src/dashboard.mjs` to restrict file access to:
- `dashboard/`
- `assets/`
- `src/docs/`
- `torch-config.json`
- Added `test/dashboard-security.test.mjs` to verify the fix and prevent regressions.
The fix ensures that sensitive files like `package.json` or source code cannot be accessed via the dashboard server.
Co-authored-by: PR0M3TH3AN <[email protected]>
