Skip to content

Security: zrack/gex-terminal

Security

SECURITY.md

Security Policy

gex-terminal handles broker and market-data credentials through local environment variables. Please do not commit credentials, account identifiers, API tokens, session tokens, or private market-data entitlement details.

Supported Versions

This project is pre-1.0. The source/package version is 0.2.0, but this repository does not claim a published PyPI artifact or release tag. Security fixes target the latest main branch.

Reporting a Vulnerability

If you find a credential-handling issue, token exposure path, or unsafe logging behavior, please do not open a public issue with sensitive details.

Instead, contact the maintainer privately through GitHub or by using GitHub's private vulnerability reporting if it is enabled for the repository.

Include:

  • A short description of the issue.
  • The affected command or data mode.
  • Steps to reproduce with sanitized inputs.
  • Whether any credential, token, or account data could be exposed.

Credential Safety

  • Keep real credentials in .env, not .env.example.
  • Confirm .env is not staged before committing.
  • Remove tokens, account IDs, and private entitlement details from logs.
  • Prefer demo or replay mode when sharing screenshots or bug reports.
  • Treat captured sessions as potentially licensed market data. The capture format stores normalized messages rather than raw provider frames, but that does not make the data redistributable.
  • Repository-local *.gex-session.jsonl captures and their .partial files are ignored by default. Review license and content before deliberately force-adding one.
  • Do not share .partial captures as evidence; they are intentionally incomplete and cannot pass integrity verification.
  • Run tradovate-certify only with the explicit --ack-live-network flag. Its report is designed to be redacted, but review any artifact before sharing it.
  • Never paste Tradovate access tokens, market-data tokens, authorization frames, usernames, client secrets, or account identifiers into issues or fixtures.

There aren't any published security advisories