Skip to content

Security: evaleval/auto-benchmarkcard

Security

SECURITY.md

Security

Auto-BenchmarkCards is research code, not a hardened public service. The pipeline follows candidate URLs from input records and fetches papers, repository pages, dataset pages, and other web sources. It does not currently provide a complete blocklist for localhost, private networks, cloud metadata endpoints, or other sensitive destinations.

Do not expose the CLI or webhook to untrusted inputs without adding URL validation, network egress controls, authentication, request limits, and logging appropriate to the deployment. Run research jobs in an isolated environment with least-privilege credentials. Never place credentials in committed files; use environment variables or an approved secret manager.

Report suspected vulnerabilities privately to aris.hofmann@ibm.com. Do not include live credentials, participant data, or sensitive source material in a public issue.

There aren't any published security advisories