Skip to content

Security: munvard/dockterm

Security

SECURITY.md

Security Policy

Reporting a vulnerability

Please report security issues privately, not in public issues or PRs.

You'll get an acknowledgement as soon as possible. Please include: affected version, platform, a description, and reproduction steps or a proof of concept. Please give a reasonable window to address the issue before any public disclosure.

Scope

DockTerm is a local desktop application. There is no DockTerm server, account system, or backend — so the threat model is local. Areas most relevant to report:

  • Escaping the project filesystem jail (path traversal, symlink escape).
  • Executing code from an opened (untrusted) project repository — e.g. via git hooks, or via the MCP/skills config readers.
  • Leaking masked secrets from the MCP/skills panels.
  • Renderer sandbox escapes, IPC validation bypasses, or remote-content loading.

What is not a vulnerability: that the terminal runs whatever you type. DockTerm grants no privilege a normal terminal doesn't; it only ensures that only its own local UI can drive that terminal.

Supported versions

This is pre-1.0 software. Security fixes target the latest release on main. There are no backports yet.

What DockTerm guarantees

  • No telemetry, analytics, or crash reporting.
  • No network calls at runtime except git operations you initiate (push/pull) and external links you click.
  • No accounts, no cloud, no stored credentials/tokens.
  • Renderer runs with contextIsolation: true, nodeIntegration: false, sandbox: true; production loads from a custom app:// origin under a strict CSP with no remote content.

See docs/SECURITY_MODEL.md for the full model.

There aren't any published security advisories