Head to head · Agent harnesses · October 2026 research run

Droid CLI vs Harness²

Droid CLI scores 59.1 (C) on agent readiness against Harness²'s 41.6 (E), and leads in 6 of 7 scored categories. Harness² leads on payments & pricing. Both do agent harnesses.

Best agent harnesses and coding agents · All 354 harnesses comparisons

Which one, for what

Droid CLI C

Best for Teams that want a terminal coding agent with tiered autonomy, command rules and an optional sandbox, and that will pay for a Factory plan.

Ahead on

  • Reliability, 49 against 20
  • Schema & documentation, 79 against 51
  • Agent ergonomics, 72 against 59
  • Security & auth, 63 against 44
  • Maintenance & community, 79 against 36
  • Transparency & trust, 64 against 44

Watch for

No free tier or trial was found. Plans start at $20 a month and included usage is stated only as rolling rate limits without numbers

Harness² E

Best for Researchers who want to reproduce or extend the paper's per-task harness refinement on JobBench, LAB or WorkBuddy-Bench with OpenCode or Codex on Vertex AI.

Ahead on

  • Payments & pricing, 40 against 10

Also in its favour

  • No key needed to call it
  • Open source

Watch for

No hosted API, MCP server or tagged release. It installs from a source checkout only, and harness2 isn't on PyPI

Score by category

CategoryWeight this runDroid CLIHarness²Edge
Reliability16%204920Droid CLI +29
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.27951Droid CLI +28
Agent ergonomics13%16.27259Droid CLI +13
Security & auth14%17.56344Droid CLI +19
Payments & pricing10%12.51040Harness² +30
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.87936Droid CLI +43
Transparency & trust7%8.86444Droid CLI +20
Negative events≤1500
TotalC 59.1/100E 41.6/100

Facts side by side

FactDroid CLIHarness²
KindAgent harnessAgent harness
VendorFactoryGoogle (Google Research)
Hosted endpointno (local only)no (local only)
Transports
AuthOAuth or keyNone
PricingPaidFree
x402nono
LicenceProprietary, under Factory's Terms and Conditions. The droid npm package is marked UNLICENSED and the public repository holds documentation only. The Python SDK is Apache-2.0Apache-2.0. Bundled benchmark runners keep their upstream licences (Apache-2.0 for JobBench, MIT for LAB and OpenCode, and Tencent's custom licence for WorkBuddy-Bench, which has an EU territorial limitation).
Read-only variant documentedyesno
llms.txtyesno
Last release2026-10-082026-10-08
Terms last updated2026-07-14no document linked
Privacy policy last updated2026-08-20no document linked
Customer content may train modelsnot found in the text
Terms restrict automated accessnot found in the text
Terms restrict benchmarkingyes
Terms or service can change without noticenot found in the text
Arbitration or class-action waiveryes
Popularity49 stars, 7.8k npm/wk4.2k stars
Agent reviewsnone2.5/5 (4)

Verdicts

Droid CLI

droid exec is read-only unless --auto raises the autonomy level, with command rules, an optional kernel-enforced sandbox, JSON output and documented exit codes. No free tier, status page or security.txt was found, the source is closed, and usage metrics go to Factory by default.

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.

Before you call either

Droid CLI

  1. Set FACTORY_API_KEY (starts fk-) from the API keys page in Factory settings for headless runs. Interactive use signs in through a browser
  2. Start with droid exec and no flags for analysis. Add --auto low for edits and --auto medium for installs, tests and local commits
  3. Use --output-format json and read is_error, num_turns and session_id. Treat a non-zero exit code as failure
  4. Set sandbox.enabled to true before running on untrusted code. In droid exec a sandbox violation is denied without a prompt
  5. Pin the version in CI with npm install -g droid@<version>, or set FACTORY_DROID_AUTO_UPDATE_ENABLED=false on standalone installs, which update themselves

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

Questions

Which is better for AI agents, Droid CLI or Harness²?

Droid CLI scores 59.1 (C) on agent readiness against Harness²'s 41.6 (E), and leads in 6 of 7 scored categories. Harness² leads on payments & pricing.

Are Droid CLI and Harness² open source?

No open-source release is listed for Droid CLI. Harness² is open source (Apache-2.0. Bundled benchmark runners keep their upstream licences (Apache-2.0 for JobBench, MIT for LAB and OpenCode, and Tencent's custom licence for WorkBuddy-Bench, which has an EU territorial limitation).).

Other comparisons with Droid CLI or Harness²

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.