harness-test-drive
Skill1.5k starsChanged 5 months agoCC-BY-SA-4.0View on GitHub
Test-drives coding agents (Claude Code, Codex, Gemini CLI) on tasks mined from the user's own git history: each agent gets a past commit message in a fresh copy of the repo, and the repo's own tests score how many tasks it completes, dollars per task, minutes, and change size. Use when the user asks which coding agent or harness works best on their codebase; wants to compare or benchmark agents on their own repository instead of trusting leaderboards such as SWE-bench; wants to trial one agent against another before switching; or asks what each agent costs per fixed bug. Spends money: the harness CLIs call their model providers, so a dollar cap is required. The scripts make no network calls.
- Installs packages
Trending
More talk lately.about the whole repository
Who's talking about RyanAlberts/best-of-Agent-Harnesses
- A Hacker News story“Best-of-Agent-Harnesses – Ranked list of 167 AI agent harnesses, rescored weekly”Hacker News · Oct 2, 2026 · 7 points
- xfey/MCP-Zero“MCP-Zero is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses)”GitHub · Jun 11, 2026 · 1 reactions
- sudosubin/agents.nix“claude-code-marketplaces.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-11-08 → …”GitHub · Nov 16, 2026
- sudosubin/agents.nix“claude-code-plugins.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-11-08 → 0-unstable-2026-11-15”GitHub · Nov 15, 2026
- sudosubin/agents.nix“agent-skills.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-11-08 → 0-unstable-2026-10-09”GitHub · Nov 10, 2026
The biggest 5 of 30 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 30
Talked about on GitHub and Hacker News (30 times).
On GitHub · 29
- Jun 11, 2026MCP-Zero is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses) — xfey/MCP-Zero1 reaction
- Nov 16, 2026claude-code-marketplaces.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-11-08 → 0-unstable-2026-11-15 — sudosubin/agents.nix0 reactions
- Nov 15, 2026claude-code-plugins.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-11-08 → 0-unstable-2026-11-15 — sudosubin/agents.nix0 reactions
- Nov 10, 2026agent-skills.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-11-08 → 0-unstable-2026-10-09 — sudosubin/agents.nix0 reactions
- Nov 10, 2026Process additional landscape items — gitaroktato/harness-engineering-technology-landscape0 reactions
24 more on GitHub
- Nov 10, 2026agent-skills.github.ryanalberts.best-of-agent-harnesses: init at 0-unstable-2026-11-08 — sudosubin/agents.nix0 reactions
- Nov 8, 2026claude-code-plugins.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-11-03 → 0-unstable-2026-11-08 — sudosubin/agents.nix0 reactions
- Nov 8, 2026claude-code-marketplaces.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-11-03 → 0-unstable-2026-11-08 — sudosubin/agents.nix0 reactions
- Nov 3, 2026claude-code-marketplaces.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-11-01 → 0-unstable-2026-11-03 — sudosubin/agents.nix0 reactions
- Nov 3, 2026claude-code-plugins.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-11-01 → 0-unstable-2026-11-03 — sudosubin/agents.nix0 reactions
- Nov 1, 2026claude-code-plugins.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-10-25 → 0-unstable-2026-11-01 — sudosubin/agents.nix0 reactions
- Nov 1, 2026RyanAlberts/best-of-Agent-Harnesses — DavidGeeraerts/AI0 reactions
- Nov 1, 2026claude-code-marketplaces.github.ryanalberts.best-of-agent-harnesses: 0-unstable-2026-10-25 → 0-unstable-2026-11-01 — sudosubin/agents.nix0 reactions
- Oct 28, 2026claude-code-plugins.github.ryanalberts.best-of-agent-harnesses: init at 0-unstable-2026-10-25 — sudosubin/agents.nix0 reactions
- Oct 28, 2026claude-code-marketplaces.github.ryanalberts.best-of-agent-harnesses: init at 0-unstable-2026-10-25 — sudosubin/agents.nix0 reactions
- Oct 8, 2026P1: Harness capability matrix and model×harness comparative evaluation — sp80808/Lattice0 reactions
- Oct 8, 2026P1: Portable skill packs and agent workflow recipes with lazy loading — sp80808/Lattice0 reactions
- Oct 8, 2026P0: Durable agent sessions and checkpoint-resume with TAP replay safety — sp80808/Lattice0 reactions
- Oct 8, 2026P0: Progressive MCP tool discovery and capability-scoped tool registry — sp80808/Lattice0 reactions · 1 comment
- Aug 12, 2026[Resource]: agent-harnesses-mcp — hesreallyhim/awesome-claude-code0 reactions · 1 comment
- Jun 11, 2026claude-mem is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses) — thedotmack/claude-mem0 reactions · 1 comment
- Jun 11, 2026AgencyBench is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses) — GAIR-NLP/AgencyBench0 reactions
- Jun 11, 2026RepoMaster is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses) — QuantaAlpha/RepoMaster0 reactions
- Jun 11, 2026AutoHarness is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses) — aiming-lab/AutoHarness0 reactions
- Jun 11, 2026claw-code-agent is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses) — HarnessLab/claw-code-agent0 reactions
- Jun 11, 2026open-harness is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses) — MaxGfeller/open-harness0 reactions
- Jun 11, 2026spring-ai-tool-search-tool is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses) — spring-ai-community/spring-ai-tool-search-tool0 reactions
- Jun 11, 2026ToolRAG is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses) — antl3x/ToolRAG0 reactions
- Jun 11, 2026ToolGen is a featured pick in Best of Agent Harnesses (curated list of 101 agent harnesses) — Reason-Wang/ToolGen0 reactions · 1 comment
On Hacker News · 1
In this repository12 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- .cursor/rules/git-author.mdcCursor rule
- llms.txtllms.txt
- agents-md-checkerSkill
- claim-checkSkill
- guardrail-testerSkill
- regression-finderSkill
- rules-to-guardsSkill
- runaway-guardSkill
- sandbox-checkSkill
- session-waste-reportSkill
- tool-design-checkerSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
Licensed with conditions — read it at the source. We count its rules but don't quote them. Read them on GitHub.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

