Do not open a public issue for vulnerabilities, leaked credentials, sandbox escapes, verifier bypasses, prompt or dataset leakage, or unsafe task assets. Email labs@benchflow.ai with:

• the affected path, version, or commit
• reproduction steps and expected impact
• whether credentials, private evaluation data, or third-party systems are at risk
• a safe contact method for follow-up

We will acknowledge a complete report within five business days and coordinate disclosure after a fix or mitigation is available.

Security-sensitive surfaces include:

• participant-provided Dockerfiles, scripts, skills, and task assets
• verifier and oracle isolation
• Daytona or Docker sandbox lifecycle
• provider and Hugging Face credentials
• private held-out evaluation data
• generated trajectories, checkpoints, and model artifacts
• protocol adapters, including the implemented OpenEnv server/client boundary

Never commit secrets, raw provider responses containing secrets, checkpoints, private eval tasks, or unreviewed job dumps. Use environment variables or a secret manager, pin external revisions, and keep generated runs in ignored directories.

The repository includes posttrainarena-train openenv-serve, which wraps BenchFlow-backed tasks behind the OpenEnv client/server protocol. Operators must keep private task snapshots and verifier credentials server-side, bind the raw service to loopback or a private network, and place any remote access behind encrypted authenticated ingress plus network allowlisting. The command has no built-in authentication and must never be exposed directly to the public internet. Preserve sandbox isolation across resets and run the lifecycle plus Docker-parity tests before changing this boundary.

The competition is a proposal and does not currently provide a production SLA.