Skip to content

Security: ReallyArtificial/otterly

Security

SECURITY.md

Security Policy

Reporting a vulnerability

If you've found a security issue in otterly, please don't open a public GitHub issue. Instead, email:

harsh.joshi.pth@gmail.com

Use a subject line that starts with [otterly-security] so it doesn't get buried. Include:

  • A description of the issue
  • Steps to reproduce
  • Affected versions (if known)
  • Your assessment of impact

You can expect an acknowledgement within 72 hours. If the issue is confirmed, a fix and a coordinated disclosure timeline will follow.

Threat model — what otterly does and doesn't protect

otterly is a local transport layer. Understanding what it does (and doesn't) helps frame what counts as a vulnerability.

In scope

  • Bugs in the HTTP / WebSocket server that allow unauthorized access to local routes (when OTTERLY_API_KEY is set)
  • Bugs that cause otterly to mis-route requests away from the intended claude CLI invocation
  • Path traversal, command injection, or similar issues in the way otterly invokes the local CLI
  • DoS vectors in the server (rate limit bypass, queue starvation, etc.)
  • Vulnerabilities introduced by the one runtime dependency (ws) that have a clear otterly-specific impact

Out of scope

  • Anything in the upstream Claude Code CLI itself — please report those to Anthropic directly
  • The user's choice to expose port 11434 to the public internet (this is documented as a bad idea; running otterly without OTTERLY_API_KEY on a public address is a configuration mistake, not a vulnerability)
  • The user's choice to share their Claude Code subscription with others via otterly (an Anthropic ToS issue, not a security one)
  • Bugs in third-party clients (OpenClaw, Cursor, Aider, etc.) that happen to use otterly

Supported versions

Only the latest 0.x release receives security fixes. Older versions are not patched.

There aren't any published security advisories