| Version | Supported |
|---|---|
| 2.x | Yes |
| < 2.0 | No |
If you discover a security vulnerability in LingFlow AI, please report it responsibly:
- Open a GitHub issue titled "Security report" without exploit details, and ask for a private channel; details are then exchanged privately
- Include a description of the impact, steps to reproduce and affected versions once in private
- Allow reasonable time for a fix before public disclosure
This is a volunteer open-source project; response times are best effort.
- API keys are stored in
chrome.storage.localwithTRUSTED_CONTEXTSaccess level - Content scripts do not have direct access to API keys and run only when the user invokes LingFlow on a page (or opts into the selection button)
- All inter-script messages are validated against a schema with size limits
- The extension requests minimal permissions
- No remote code execution — all code is bundled locally