DeepSeek Harness (dsh) vs Harness²

DeepSeek Harness (dsh)

Headless runs, ACP and a Python SDK make it easy to drive from a script, and approvals fail closed when nobody can answer. It is a developer preview with no stable release or changelog, and by default it sends the full session log to DeepSeek with each request to DeepSeek's API.

Harness²

Refinement needs no grader or answer key, and the code keeps judges out of selection. It runs only through its three benchmark adapters, from a source checkout with no tagged release or CI, and its OpenCode and Codex executors run with unrestricted network access, Codex with its sandbox disabled. The benchmark gains are the authors' claims.

DeepSeek Harness (dsh)

1. Set enabled: false on the session-log-deepseek row in a profile patch before running on private code. Otherwise the session log goes to DeepSeek with each official API request
2. Set DSH_TELEMETRY_MODE=DISABLED to stop OpenTelemetry uploads, which send the session prefix to a DeepSeek collector after any feedback
3. Treat exit code 1 and the turn_end reason as failure. A --json stream can end with a final event for a failed run
4. Run in a disposable container or VM. The sandbox doesn't confine reads or network, and SAFETY.md says the project has had no security audit
5. Pin the version. npm latest (0.2.0-rc.2) and next (0.2.1-rc.2) differ, DeepSeek's changelog asks for 0.2.1-rc.1 or later, and the README warns of breaking changes

Harness²

1. Don't plan to call Harness² as a service. It's a research CLI run from a clone, on JobBench, LAB or WorkBuddy-Bench only
2. Ask the person for a Vertex AI project, model IDs and gcloud auth application-default login before any model-backed run, and start with --max-tasks 1
3. Run executors behind an egress firewall and block huggingface.co and github.com after setup, as the docs advise, so rubrics can't be fetched
4. Write .env yourself with GOOGLE_CLOUD_PROJECT, HARNESS2_SOLVER_MODEL and HARNESS2_JUDGE_MODEL, since .env.example isn't shipped, and never commit it
5. Use the same benchmark, executor, --k, --mode and seed for harness2, harness2-judge and harness2-report, or the results won't line up