Best of Agent Harnesses

Browse all 188

Categories

Pick by use case

• I want a turnkey coding agent today — opencode, Cline, Codex, Gemini CLI, OpenHands
• I want an always-on personal agent that lives in my chat apps — OpenClaw, Hermes, Khoj, Agent Zero, OpenHarness (HKUDS)
• I want to extend Claude Code, Codex, or OpenCode with skills and slash commands — Anthropic Skills, wshobson/agents, superpowers, GStack, pmstack
• I want to build my own coding harness from scratch — Claude Agent SDK, Google ADK, AutoHarness, SWE-agent, RepoMaster
• I want a drop-in memory layer for agents — Mem0, Graphiti (Zep), claude-mem, agentlog, letta
• I want to plug hundreds to thousands of tools without context bloat — MCP-Zero, ToolGen, ToolRAG, langgraph-bigtool
• I want multi-agent orchestration — openai-agents-python, crewAI, autogen, Microsoft Agent Framework, PraisonAI
• I want a general LLM app framework — langgraph, langchain, llama-index, pydantic-ai, agno
• I want low-code / visual workflows — langflow, Dify, n8n
• I want browser-using agents — browser-use, Stagehand, WebVoyager, puppeteer-real-browser-mcp
• I want sandboxed code execution for agent-generated code — E2B, Agent Sandbox, smolagents, OpenHands
• I want to evaluate or benchmark agents — SWE-bench, Terminal-Bench, AgencyBench, inspect_ai, WebArena
• I want a deep research / autonomous research agent — deepagents, gpt-researcher, openagents
• I want a provider-agnostic LLM pipe (not a framework) — LiteLLM, vercel/ai

Decision guides

• Agent evals: SWE-bench vs inspect_ai vs Terminal-Bench — You changed your agent: new model, new prompt, new tools. Did it get better or worse? An eval is how you answer that with a number instead o
• The best AI agent harnesses in 2026, ranked by category — These are the top three agent harnesses in each of the 12 categories of best-of-Agent-Harnesses, a hand-curated list of 188 harnesses re-ran
• Browser agents: browser-use vs Stagehand vs Playwright MCP vs chrome-devtools-mcp — "Browser agent" covers three different kinds of product, and most bad picks here come from comparing across the lanes instead of within one.
• Browser infrastructure for agents: Browser Use vs Browserbase vs Steel vs Hyperbrowser — Agent libraries like browser-use and Stagehand decide what to click; something still has to run the browsers they click in. At small scale t
• Claude Code skill packs: superpowers vs GStack vs get-shit-done vs Anthropic Skills — A skill is a folder of instructions (a SKILL.md file, plus any scripts it needs) that a coding agent loads only when the task matches, inste
• Eval and observability platforms: Langfuse vs LangSmith vs Braintrust vs Phoenix — Benchmarks tell you how a model ranks; your production agent still fails in ways no public exam covers. An eval and observability platform i
• Harness engineering: what it is, and the tools for each practice — Harness engineering means building everything around an AI model that turns it into a dependable agent: what it sees, which tools it gets, w
• How to pick a harness — This is the decision guide for [best-of-Agent-Harnesses](../README.md), a curated, ranked list of the runtimes that turn an AI model into a
• How to test-drive a harness — Spec sheets cannot answer "which harness should I use," because an agent's performance is a property of the *pairing* between harness and mo
• Managed vs self-hosted always-on agents — An always-on agent keeps working between your messages, and in 2026 you can rent one (Grok Bot, Claude Managed Agents), run one for your who
• Agent memory layers: Mem0 vs Zep vs Letta vs claude-mem — Agents forget. A model keeps nothing between sessions, so anything your agent should still know tomorrow (who the user is, what was decided,
• Multi-agent orchestration: OpenAI Agents SDK vs CrewAI vs AutoGen vs Agent Framework vs LangGraph — Orchestration is the layer that coordinates several AI agents working on one job: who acts next, what they share, and what happens when a st
• OpenClaw vs Hermes: the always-on personal-agent debate — An always-on personal agent is a program that runs all day on your own machine, talks to you through the chat apps you already use (WhatsApp
• Context files for agents: AGENTS.md vs CLAUDE.md vs skills vs MCP tool search — A model has a context window: a fixed amount of text it can consider at once. Everything competes for that space: your instructions, the def
• Agent sandboxing: what it is and how to pick — An AI agent does not just suggest code. It runs code, opens web pages, and edits files on a real computer. Agent sandboxing means making tha
• Terminal coding agents: opencode vs Codex vs Gemini CLI vs crush vs goose — The most-asked pick in this list: *"I want a turnkey coding agent in my terminal today."* A terminal coding agent is a program you run in yo
• Why the harness matters more than the model — The same model weights score about 30% on ARC-AGI-3 as a bare model and 95.5% inside a good harness, and that gap is the number agent builde

Templates

Files you copy into your project to set a harness up safely.

• One AGENTS.md for every coding agent — One briefing file that Codex, Claude Code, Cursor, OpenCode, GitHub Copilot, Gemini CLI, and Aider all read, so you write your build command
• Safe Claude Code settings: permissions and a guard hook — Project settings that stop Claude Code from force-pushing, wiping work, reading secrets, or piping downloads into a shell, while leaving eve
• Minimal agent harness in Python — A working coding-agent harness in about 180 lines of Python: the loop, three tools, a permission gate, a context file, a turn budget, and a

Playbooks

Step-by-step guides, one task each.

• Build your own agent harness — Build a working coding agent in about an hour: a loop, three tools, permissions, a context file, a budget, and crash recovery, in one Python
• Write one AGENTS.md for every coding agent — Write one briefing file that Codex, Claude Code, Cursor, OpenCode, Copilot, Gemini CLI, and Aider all read, test that each tool actually loa

Skills

Folders you install into your agent. Each one runs a script on your own setup and hands back a result you can check.

• Check which instruction files your coding agents load: See which instruction files Claude Code, Codex, Gemini CLI, OpenCode, Cursor, GitHub Copilot, and Aider load from your repo, what each one c
• Check your agent's 'tests pass' claims: Find out how often your coding agent said tests passed without a passing run to back it up, and whether the current change quietly weakened
• Test your agent's guardrails against dangerous commands: Checks whether the permission rules and hooks you already have in Claude Code, Codex, Gemini CLI, OpenCode, or Cursor stop about 90 dangerou
• Test-drive coding agents on tasks from your own repo: Runs Claude Code, Codex, and Gemini CLI on tasks taken from your own git history, scores each one with your repository's tests, and reports
• Find the update where your coding agent started working differently: Splits your own Claude Code or Codex sessions by version, model, or week, measures how the agent works in each, and places the change at the
• Turn the rules your agent breaks into hooks: Finds the rules in AGENTS.md, CLAUDE.md, and GEMINI.md that your coding agent breaks, counts every break in your recent sessions, and turns
• Stop a runaway agent session: A hook for Claude Code and Codex that stops a live session when the agent repeats the same call, keeps failing, or spends past a dollar cap
• Check what your coding agent can reach: A one-minute probe that runs through your agent's own shell and shows which secret files, trust files, Docker, and sudo access it really has
• See where your coding agent wastes tokens and money: A report on your own Claude Code, Codex, Gemini CLI, and OpenCode sessions that shows which habits cost the most, such as re-read files, hug
• Grade your MCP tools the way a model reads them: Point it at an MCP server or a saved tool list to grade each tool's purpose, parameters, annotations, and token size, or let it find the MCP

FAQ

• Which agent harnesses can run unattended (headless)?
• Which agent harnesses survive a crash mid-task (durable)?
• How many of these agent harnesses are open source?
• What is an agent harness?
• Does the harness matter more than the model?
• Is Grok Bot an agent harness?
• How is this list ranked?
• How can an AI agent use this list directly?

All questions →

For agents

This list is published machine-readable so coding and research agents can recommend harnesses directly:

• harnesses.json — every project with tier, tags, axes, license, example, and use-case index.
• llms.txt — the whole list in one agent-readable file.
• llms-full.txt: llms.txt plus the full text of every decision guide, one file.
• harnesses.jsonld — schema.org Dataset + ItemList.
• feed.json — JSON Feed of refreshes.
• MCP server: uvx agent-harnesses-mcp — pick_harness, search_harnesses, get_harness, comparisons.
• Skills: npx skills add RyanAlberts/best-of-Agent-Harnesses lists every skill; the MCP server's list_skills and get_skill return them too.