Skip to content

Security: INNO-HI/ClaudeUsageWidget

Security

SECURITY.md

Security Policy

Supported Versions

Only the latest minor release receives security fixes.

Version Supported
1.1.x
1.0.x
< 1.0

Reporting a Vulnerability

Please do not file a public issue for security vulnerabilities.

Instead, report privately to board@innohi.ai.kr with the subject [SECURITY] Claude Usage Widget.

Include:

  • A description of the vulnerability
  • Steps to reproduce
  • The version(s) affected
  • Any proof-of-concept code (if applicable)
  • Your name/handle for credit (optional)

We will:

  1. Acknowledge receipt within 3 business days
  2. Confirm or dispute the issue within 7 business days
  3. Issue a patched release as soon as practical (typically within 30 days for high-severity issues)
  4. Credit the reporter in the release notes (unless they prefer to remain anonymous)

Out of Scope

  • Issues in dependencies (please report upstream)
  • Bugs that require physical access to the user's unlocked Mac
  • Social engineering attacks against the maintainer
  • Theoretical risks without a working PoC

Threat Model

This widget reads OAuth credentials from ~/.claude/.credentials.json (created by Claude Code itself) and queries https://api.anthropic.com/api/oauth/usage. It does not:

  • Send messages to Claude (zero token cost)
  • Transmit data to any third party
  • Persist credentials beyond the file Claude Code already maintains
  • Run any privileged operations (no sudo, no system file modification)

The auto-update mechanism uses Sparkle 2.x with EdDSA signature verification — updates are cryptographically verified before installation.

There aren't any published security advisories