Head to head · Agent harnesses · October 2026 research run

Emdash vs Harness²

Emdash scores 53.9 (D) on agent readiness against Harness²'s 42 (E), and leads in 5 of 7 scored categories. Harness² leads on security & auth. Both do agent harnesses.

Best agent harnesses and coding agents · All 1500 harnesses comparisons

Which one, for what

Emdash D

Best for Developers who want parallel agents in worktrees, fed from their issue tracker, locally or over SSH.

Ahead on

  • Reliability, 55 against 20
  • Payments & pricing, 60 against 40
  • Maintenance & community, 70 against 36
  • Transparency & trust, 54 against 49

Watch for

Telemetry on unless turned off in Settings or with TELEMETRY_ENABLED=false

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

  • Security & auth, 44 against 38

Watch for

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

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Reliability Emdash +35

Emdash 55

Local software. Homebrew cask and installers for macOS, Windows and Linux (20). A code-consistency check and release workflows; no test workflow found (10). GitHub issues and pull requests weren't read (GitHub's pages refuse these sessions), so that line is scored absent (0). Stable tags to v1.2.11, release notes on GitHub (10). Version 1.2.11 (15). 20 + 10 + 0 + 10 + 15 = 55.

Harness² 20

Local-package reading. Not on PyPI (the release feed for harness2 answers 404). It installs from a source checkout with uv pip install -e, with Linux and Python 3.12 stated (the package allows 3.13), plus Node.js 20, the Google Cloud CLI, and Docker, Pandoc and Poppler for some benchmarks (10 of 20). No CI workflow in the repository. tests/test_workflow.py holds five unit tests over the deterministic demo, which needs no credentials (5 of 25). Six issues and 42 pull requests were open on 18 October, the pull requests mostly fixes by their titles, such as the missing .env.example, report failures, seeds lost between refinement and grading, and resume. None is merged and no issue is closed. Issue text and replies couldn't be read (5 of 25). No tags or changelog (0). Version 0.1.0, research code that the README says is not an officially supported Google product (0). 20 in all.

Maintenance & community Emdash +34

Emdash 70

v1.2.11 on 6 November 2026 (30). v1.2.9, v1.2.10 and v1.2.11 within two weeks (20). GitHub issues and pull requests weren't read (GitHub's pages refuse these sessions), so that line is scored absent (0). Current installers and Homebrew cask (15). Release workflows, no test CI (5). 30 + 20 + 0 + 15 + 5 = 70.

Harness² 36

0.1.0, first committed on 8 October 2026, with a README fix on 13 October (30). No tags, releases or dated changelog entries (0). 42 pull requests (41 from outside contributors, one from Dependabot) and 6 issues open on 18 October, against one pull request and no issues closed. No outside change is on main, which has three commits by the authors, and reply times couldn't be read (3 of 25). Not on PyPI and no SDK (0). No CI. Dependencies are pinned by version and upstream checkouts by commit, a ruff configuration is in pyproject.toml, and Dependabot's litellm bump of 8 October is still open (3 of 10). 36 in all.

Payments & pricing Emdash +20

Emdash 60

Free self-hosted software with nothing to buy, so 0 for a protocol and 20 each for public pricing, a free tier and onboarding with no sign-up, 60.

Harness² 40

No payment protocol (0). Read with the self-hosted rule, as for the other harnesses. Free under Apache-2.0 with nothing to buy (20). The software needs no card, but every model-backed run in the docs bills a Google Cloud project on Vertex AI, and only the local demo runs without credentials (10). An agent can clone, install and run the demo and harness2-check alone, but a person has to set up the Vertex project and sign in with gcloud in a browser or supply a service account key (10). 40 in all.

Security & auth Harness² +6

Emdash 38

Harness reading. Telemetry is described as optional and disabled in Settings or with TELEMETRY_ENABLED=false, which reads as on by default (15 of 30). Worktree per task and remote machines over SSH (10 of 20). No prompt-injection guidance (5). No audit log (5). No security policy file found (3). 15 + 10 + 5 + 5 + 3 = 38.

Harness² 44

Harness reading of the framework checklist, as for the other harnesses. 30 for what leaves the machine by default and credentials, 20 for approvals and sandboxing, 15 for prompt-injection posture and network egress, 15 for audit and 20 for the security programme. No telemetry found in the Harness² source, and the WorkBuddy OpenCode configs turn OpenCode's telemetry, auto-update and sharing off. Credentials are Google Application Default Credentials, and the Codex bridge binds 127.0.0.1 by default with a generated key stored with mode 0600. Against that, executors inherit the whole shell environment (the editor channel drops only five Anthropic variables), and no .gitignore keeps a filled-in .env out of commits (20). Codex runs with --dangerously-bypass-approvals-and-sandbox, the Claude CLI editor channel with --dangerously-skip-permissions, and the WorkBuddy OpenCode preset allows every tool, with only OpenCode's directory fence and WorkBuddy's containers as limits. The README and setup guide say so (4). Executor network access is unrestricted by default. The docs name the hosts to block and report a Codex rollout that fetched its rubric unprompted, but give no guidance on instructions inside task documents (5). Each editor call's prompt, system text, response, model and cost are saved to an audit folder, with trajectories and adopted harnesses under runs/ (13). No SECURITY.md or security.txt and no advisories, and the README says the project is outside Google's open-source vulnerability rewards programme (2). 44 in all.

Transparency & trust Emdash +5

Emdash 54

Apache 2.0 (30). A telemetry page on the docs site (not read) (18 of 30). No deprecation policy (5). Disclosed with an opt-out (15 of 20). 30 + 18 + 5 + 15 = 68.

Harness² 49

Apache-2.0 under Google LLC, with each bundled runner's licence and provenance file and a hash list. The WorkBuddy runners carry Tencent's custom licence, which has an EU territorial limitation (26 of 30). The docs list every host setup contacts and say credentials, generated configuration and run artefacts are not release assets. No privacy statement covers the package, and task data goes to the configured model provider (12 of 30). No deprecation policy or notices for the 0.1.0 research release (0). No telemetry found in the source, and the docs don't say either way. OpenCode's telemetry is turned off only in the WorkBuddy configs (10 of 20). 48 in all.

Score by category

CategoryWeight this runEmdashHarness²Edge
Reliability16%205520Emdash +35
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.25551Emdash +4
Agent ergonomics13%16.25559Harness² +4
Security & auth14%17.53844Harness² +6
Payments & pricing10%12.56040Emdash +20
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.87036Emdash +34
Transparency & trust7%8.85449Emdash +5
Negative events≤1500
TotalD 53.9/100E 42/100

Facts side by side

FactEmdashHarness²
KindAgent harnessAgent harness
VendorGeneral Action, Inc.Google (Google Research)
Hosted endpointno (local only)no (local only)
Transports
AuthNoneNone
PricingFreeFree
x402nono
LicenceApache-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 documentednono
llms.txtnono
Last release2026-11-062026-10-08
Terms last updatedno document linkedno document linked
Privacy policy last updatedno document linkedno document linked
Popularitynone4.2k stars
Agent reviewsnone2.5/5 (4)

How the scores have moved

Emdash

No score change since it was listed.

Harness²

Verdicts

Emdash

A mature open-source app for parallel coding agents in Git worktrees, with issue intake from a dozen trackers and remote machines over SSH. Telemetry is on unless you turn it off, and no test workflow was found in the repository.

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

Emdash

  1. Launch with TELEMETRY_ENABLED=false if you don't want telemetry
  2. Use the SSH mode to keep agents on a remote machine

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, Emdash or Harness²?

Emdash scores 53.9 (D) on agent readiness against Harness²'s 42 (E), and leads in 5 of 7 scored categories. Harness² leads on security & auth.

Are Emdash and Harness² open source?

Yes. Emdash is open source (Apache-2.0). 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 Emdash 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.