| Version | Supported |
|---|---|
| 0.5.x | Yes |
| 0.4.x | Security fixes only |
| < 0.4 | No |
Preferred: Use GitHub Security Advisories to report vulnerabilities privately.
Alternative: Email security concerns to the repository owner via GitHub profile.
| Stage | Timeline |
|---|---|
| Acknowledge receipt | Within 48 hours |
| Initial assessment | Within 7 days |
| Fix released | Within 30 days (critical: 7 days) |
- Description of the vulnerability
- Steps to reproduce
- Potential impact
- Suggested fix (if any)
cc-lens is a localhost-only dashboard that reads Claude Code session data from ~/.claude/. The security model relies on:
- TCP binding to
127.0.0.1(primary boundary) - Middleware two-layer Host header check (defense-in-depth)
- Input validation on all API parameters (isValidSlug, path traversal guards)
- Path traversal in API routes or readers (slug, session ID, file paths)
- XSS via session data rendered in the dashboard
- Cache corruption or poisoning
- Information disclosure beyond
~/.claude/directory - Bypass of localhost-only restriction
- Denial of service via malformed JSONL input
- Physical access to the machine (localhost tool assumes trusted local user)
- Social engineering
- Issues requiring non-default
CC_LENS_HOSTconfiguration (user explicitly opted in) - Vulnerabilities in upstream dependencies with no cc-lens-specific exploit path
Responsible reporters will be credited in the release notes unless they prefer to remain anonymous.
For a detailed threat model covering attack surfaces, trust boundaries, and mitigations, see docs/threat-model.md.