Skip to content

Security: Parakh-Shinde/GCP-IAMGraph

Security

SECURITY.md

Security Policy

Supported versions

GCP IAMGraph is currently an early-stage security research project.

Version Supported
0.1.x Yes
Older versions No

Security fixes are applied to the latest release and the default branch.

Reporting a vulnerability

Do not open a public GitHub issue for a vulnerability that could place users, credentials, IAM data, or cloud environments at risk.

Report security issues privately through:

Include the following information when possible:

  • affected version or commit;
  • vulnerable component;
  • steps required to reproduce the issue;
  • expected and actual behavior;
  • potential security impact;
  • logs, screenshots, or proof-of-concept details;
  • suggested remediation, if known.

Sensitive credentials, real IAM exports, access tokens, service-account keys, and confidential organization information must not be included in the report.

Response process

After receiving a report, the maintainer will:

  1. review and validate the reported behavior;
  2. determine its impact and affected versions;
  3. develop and test an appropriate remediation;
  4. coordinate disclosure with the reporter when necessary;
  5. publish a security advisory or patched release when appropriate.

This is a student-maintained open-source project, so response times may vary. Valid reports will be handled as promptly as reasonably possible.

Credential and data safety

GCP IAMGraph does not require live Google Cloud credentials for its included examples.

Before using or sharing exported IAM data:

  • remove credentials and secrets;
  • sanitize organization, folder, and project identifiers;
  • replace real user and service-account names;
  • exclude access tokens and service-account keys;
  • avoid committing confidential IAM policies;
  • review generated reports before publishing them.

Project security boundaries

GCP IAMGraph performs read-only, offline analysis. It does not:

  • modify Google Cloud IAM policies;
  • create or delete service-account keys;
  • exploit detected attack paths;
  • authenticate to Google Cloud automatically;
  • collect or transmit credentials;
  • remediate findings automatically.

Findings should be reviewed by a qualified cloud-security practitioner before production changes are made.

Responsible disclosure

Please allow reasonable time for validation and remediation before publicly disclosing an unresolved vulnerability. Good-faith security research and responsible disclosure are appreciated.

There aren't any published security advisories