Skip to content

Security: HotaruProject/Hotaru

SECURITY.md

Security policy

Supported versions

Security updates are provided exclusively for the latest commit on the main branch. Older commits and unmerged development branches do not receive backported security patches.

Version / Branch Supported
main Yes
develop Active development only
Older releases No

Reporting a vulnerability

Do not report security vulnerabilities in public GitHub issues, discussions, or Telegram chats.

Use GitHub Private Vulnerability Reporting to submit findings securely:

Submit a Security Advisory

If you cannot use GitHub Advisories, contact the project maintainers directly with PGP-encrypted details.

What to include in your report

To help us verify and resolve the issue quickly, include:

  • A clear description of the vulnerability and its potential impact.
  • Step-by-step reproduction instructions or a minimal proof of concept script.
  • The environment details (OS, Python version, Docker or host installation).
  • Any proposed remediation or mitigation steps, if available.

Vulnerability scope

High-priority issues for Hotaru include:

  • Sandbox and namespace escapes: breaking out of .hmod worker isolation or accessing the host filesystem outside the sandbox directory.
  • Credential leaks: unauthorized exfiltration of master vault keys, Telegram session strings, or auth tokens.
  • Privilege escalation: bypassing the capability model (mt, state, net) or triggering unauthorized RPC methods.
  • Sandbox firewall circumvention: defeating network blocking or IPC filters for restricted modules.

Out of scope:

  • Attacks requiring existing root access or host compromise.
  • Denial-of-service against third-party network providers.
  • Account restrictions or bans applied by Telegram for spam or terms violations.

Response timeline

  • Acknowledgment: within 48 hours of submission.
  • Assessment: initial severity rating and reproduction within 5 business days.
  • Patch and release: coordinated disclosure schedule agreed with the reporter once a fix is verified.

There aren't any published security advisories