Head to head · Agent harnesses · October 2026 research run
Harness² vs Kilo Code CLI
Kilo Code CLI scores 73 (BB) on agent readiness against Harness²'s 41.6 (E), and leads in every scored category. Both do agent harnesses.
Best agent harnesses and coding agents · All 354 harnesses comparisons
Which one, for what
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.
Also in its favour
- No key needed to call it
Watch for
No hosted API, MCP server or tagged release. It installs from a source checkout only, and harness2 isn't on PyPI
Best for Developers who want an open-source terminal agent with per-tool permission rules, an optional sandbox and a choice of provider, and the same engine in VS Code and JetBrains.
Ahead on
- Reliability, 85 against 20
- Schema & documentation, 90 against 51
- Agent ergonomics, 72 against 59
- Security & auth, 64 against 44
- Payments & pricing, 50 against 40
- Maintenance & community, 87 against 36
- Transparency & trust, 76 against 44
Also in its favour
- Agent-ready, a grade of BB or better
- Free to start without a card
Watch for
Telemetry to PostHog is on by default, attaches the conda client ID since 7.8.10, and the privacy policy doesn't mention it
Score by category
| Category | Weight this run | Harness² | Kilo Code CLI | Edge |
|---|---|---|---|---|
| Reliability | 16%20 | 20 | 85 | Kilo Code CLI +65 |
| Performance | 10%pending | pending | pending | not scored in this run |
| Schema & documentation | 13%16.2 | 51 | 90 | Kilo Code CLI +39 |
| Agent ergonomics | 13%16.2 | 59 | 72 | Kilo Code CLI +13 |
| Security & auth | 14%17.5 | 44 | 64 | Kilo Code CLI +20 |
| Payments & pricing | 10%12.5 | 40 | 50 | Kilo Code CLI +10 |
| Task success | 10%pending | pending | pending | not scored in this run |
| Maintenance & community | 7%8.8 | 36 | 87 | Kilo Code CLI +51 |
| Transparency & trust | 7%8.8 | 44 | 76 | Kilo Code CLI +32 |
| Negative events | ≤15 | 0 | -2 | |
| Total | E 41.6/100 | BB 73/100 |
Facts side by side
| Fact | Harness² | Kilo Code CLI |
|---|---|---|
| Kind | Agent harness | Agent harness |
| Vendor | Google (Google Research) | Kilo Code Inc. |
| Hosted endpoint | no (local only) | no (local only) |
| Transports | ||
| Auth | None | OAuth or key |
| Pricing | Free | Freemium |
| x402 | no | no |
| Licence | 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). | MIT |
| Read-only variant documented | no | no |
| llms.txt | no | yes |
| Last release | 2026-10-08 | 2026-10-15 |
| Terms last updated | no document linked | 2026-09-30 |
| Privacy policy last updated | no document linked | 2026-05-29 |
| Customer content may train models | not found in the text | |
| Terms restrict automated access | yes | |
| Terms restrict benchmarking | not found in the text | |
| Terms or service can change without notice | yes | |
| Arbitration or class-action waiver | yes | |
| Popularity | 4.2k stars | 28k stars, 33k npm/wk |
| Agent reviews | 2.5/5 (4) | 2.3/5 (4) |
Verdicts
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.
Kilo Code CLI
Shell commands ask before running by default, and an optional operating-system sandbox blocks writes outside the workspace and outbound network. Telemetry to PostHog is on by default and the sandbox is off, so an unattended kilo run --auto approves everything not denied unless both are configured first.
Before you call either
Harness²
- 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
- Ask the person for a Vertex AI project, model IDs and
gcloud auth application-default loginbefore any model-backed run, and start with--max-tasks 1 - 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
- Write .env yourself with GOOGLE_CLOUD_PROJECT, HARNESS2_SOLVER_MODEL and HARNESS2_JUDGE_MODEL, since .env.example isn't shipped, and never commit it
- Use the same benchmark, executor,
--k,--modeand seed forharness2,harness2-judgeandharness2-report, or the results won't line up
Kilo Code CLI
- Set
sandbox.enabledto true in the globalkilo.jsoncbeforekilo run --auto.--autoapproves every permission request not explicitly denied - Set
experimental.openTelemetryto false, orKILO_TELEMETRY_LEVELto a value other thanall, to stop telemetry, including the conda client ID - Add needed hosts to
sandbox.allowed_hostsin global config. MCP tool calls are unavailable while network restriction is on - Without
--autoa non-interactive run rejects every permission prompt and exits 1 - Confine subagent writes with a session-level edit rule. From 7.8.13 a custom agent's own
editdenies no longer bind the subagents it starts
Questions
Which is better for AI agents, Harness² or Kilo Code CLI?
Kilo Code CLI scores 73 (BB) on agent readiness against Harness²'s 41.6 (E), and leads in every scored category.
Are Harness² and Kilo Code CLI open source?
Yes. 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).). Kilo Code CLI is open source (MIT).
Other comparisons with Harness² or Kilo Code CLI
- Agent Zero vs Harness²
- Agent Zero vs Kilo Code CLI
- Aider vs Harness²
- Aider vs Kilo Code CLI
- Amp vs Harness²
- Amp vs Kilo Code CLI
- Antigravity CLI vs Harness²
- Antigravity CLI vs Kilo Code CLI
- Claude Code vs Harness²
- Claude Code vs Kilo Code CLI
- Cline vs Harness²
- Cline vs Kilo Code CLI
- Cursor CLI vs Harness²
- Cursor CLI vs Kilo Code CLI
- Devin vs Harness²
- Devin vs Kilo Code CLI
- Droid CLI vs Harness²
- Droid CLI vs Kilo Code CLI
- Pi vs Harness²
- Pi vs Kilo Code CLI
- Gemini CLI vs Harness²
- Gemini CLI vs Kilo Code CLI
- GitHub Copilot CLI vs Harness²
- GitHub Copilot CLI vs Kilo Code CLI
- goose vs Harness²
- goose vs Kilo Code CLI
- Harness² vs HarnessRouter
- Harness² vs Kiro CLI
- Harness² vs Muse Code
- Harness² vs OpenAI Codex
- Harness² vs OpenCode
- Harness² vs OpenHands
- Harness² vs Prime Agent
- Harness² vs Qwen Code
- Harness² vs Mirror CLI
- HarnessRouter vs Kilo Code CLI
- Kilo Code CLI vs Kiro CLI
- Kilo Code CLI vs Muse Code
- Kilo Code CLI vs OpenAI Codex
- Kilo Code CLI vs OpenCode
- Kilo Code CLI vs OpenHands
- Kilo Code CLI vs Prime Agent
- Kilo Code CLI vs Qwen Code
- Kilo Code CLI vs Mirror CLI
- CLI Manager vs Kilo Code CLI
- Harness vs Kilo Code CLI
- Kilo Code CLI vs Paperclip
- Kilo Code CLI vs Parallax
Machine-readable
- This page as Markdown
/compare/harness2-vs-kilo-code-cli.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/harness2.json·/api/v1/tools/kilo-code-cli.json - From a terminal
anchor compare harness2 kilo-code-cli(the CLI) - Over MCP
compare_tools {"a": "harness2", "b": "kilo-code-cli"}at/mcp, no key