| Version | Supported |
|---|---|
| 0.2.x | Yes |
| < 0.2 | No |
If you discover a security vulnerability in bq-entity-resolution, please report it responsibly.
Do NOT open a public GitHub issue for security vulnerabilities.
- Email your findings to the maintainers via the contact listed in
pyproject.toml - Include a description of the vulnerability and steps to reproduce
- If possible, include a proof-of-concept or test case
- Acknowledgment: Within 48 hours of your report
- Assessment: Within 7 days we will assess severity and impact
- Fix timeline: Critical vulnerabilities will be patched within 14 days; lower severity within 30 days
- Disclosure: We will coordinate disclosure timing with you
The following are in scope:
- SQL injection via config YAML values that bypass validation
- PII leakage in logs, error messages, or audit trails
- Authentication/authorization bypass in BigQuery operations
- Dependency vulnerabilities in direct dependencies
- Denial of service via resource exhaustion (unbounded loops, memory)
The following are out of scope:
- Vulnerabilities in BigQuery itself or Google Cloud infrastructure
- Issues requiring physical access to the deployment environment
- Social engineering attacks
- Vulnerabilities in optional development dependencies
- All user-supplied values are validated through
sql/utils.pybefore SQL interpolation - Config-level validation via Pydantic v2 field validators catches injection at load time
- Builder-level defense-in-depth re-validates at SQL generation time
- PII is redacted from SQL audit logs and exception messages
- Distributed locking uses fencing tokens to prevent stale-writer corruption
- Circuit breaker prevents cascading failures from BigQuery outages