Skip to content

Security: openclaw/libterminal

Security

SECURITY.md

Security Policy

If you believe you found a security issue in @openclaw/libterminal, report it privately.

Reporting

Open a private report through GitHub Security Advisories or email security@openclaw.ai.

Include:

  1. affected version or commit
  2. runtime and operating system
  3. minimal reproduction
  4. demonstrated impact
  5. suggested remediation, if known

Do not open a public issue until maintainers have coordinated disclosure.

Scope

Security issues in scope generally include:

  • protocol parsing or framing flaws that cross a documented trust boundary
  • terminal data leaking between sessions, streams, or consumers
  • unsafe handling of terminal control data that produces concrete impact
  • browser, Node.js, or Worker behavior differences that bypass a documented safety invariant
  • dependency or release-pipeline compromise affecting the published package

Reports must demonstrate a concrete boundary bypass or impact. Prompt injection, malicious terminal output, or hostile child-process behavior alone is not a vulnerability unless the library promises and fails to enforce a relevant boundary.

Operational Guidance

  • Keep @openclaw/libterminal, Node.js, and browser runtimes current.
  • Treat terminal input and output as untrusted data.
  • Do not log credentials or sensitive terminal contents.
  • Pin and review dependency updates before publishing.

There is currently no paid bug bounty program.

There aren't any published security advisories