Harness Engineering

Better Harness is an open-source Harness Engineering platform for coding agents. It analyzes project and, where supported, session evidence around an agent's work rather than only its final code diff, evaluating the Agent Work Loop for issues such as task understanding, validation, delivery control, and retained lessons. It turns supported gaps into prioritized findings linked to their evidence, expected outcomes, repair boundaries, and acceptance checks, and produces host-specific reports in formats including HTML, paired Markdown, or native Canvas reports. The platform also supports defining harnesses as code, running controlled experiments, inspecting evidence, and comparing outcomes across coding-agent hosts such as Claude Code, Codex, Qoder, Cursor, and GitHub Copilot CLI.

HarnessRouter Community Edition is the open-source edition of HarnessRouter, an agent-harness infrastructure that provides a unified Agent API to integrate language models and agent runtimes. It is maintained by HarnessRouter and is intended for deployment on users' own infrastructure to connect models such as Codex, Claude Code, and Hermes.

FrontierHarness Eval is an open evaluation benchmark and repository for measuring how coding-agent harness configurations affect software-engineering task results when the underlying model and runtime are held constant. It publishes benchmark definitions, task instructions and metadata, harness-version metadata, normalized aggregate and task-level results, and an agent-neutral workflow for evaluating additional harnesses. The benchmark runs the same model through multiple harness configurations on 30 tasks, using a frozen golden checkpoint and a fresh restore for each task. Its scripts provision the environment, run trials, record verifier-based pass/fail outcomes, measure cost, cache behavior, and speed, then normalize results and generate charts and reports. The repository also provides a CLI and skill-based workflow for evaluating a third-party harness, with infrastructure failures marked separately from task failures. The project focuses on terminal-based software-engineering tasks and states that its results may not generalize to other knowledge-work domains. The repository is published under the frontier-harness-eval organization, with isolated runtimes and checkpoint restores provided by Runta.

macOS Harness is an experimental, MIT-licensed, macOS-only Python harness from browser-use that gives an LLM or coding agent raw primitives for controlling a Mac in one persistent Python process, without app-specific tools, recipes, or framework rails. Its six core primitives—`see`, `key`, `type`, `click`, `ax`, and `script`—capture native and Electron app windows, send keyboard and coordinate input directly to an application process, expose Apple Accessibility and Apple Events, execute AppleScript, and draw a click-through pointer without moving the physical cursor. The agent can write missing task logic in ordinary Python during execution. The same process provides access to a real, logged-in Chrome browser through Browser Harness, as well as the local filesystem and shell commands through Python, `Path`, and `subprocess`. It can capture background application windows without bringing them to the foreground and does not activate or raise the target app. The project includes installation and skill-registration workflows, a `doctor` command for checking required macOS permissions, and verification instructions. Anonymous telemetry is enabled by default and records the CLI command category, success, duration, package version, operating system and architecture, and detected agent client; the project states that it does not record prompts, app names, screenshots, UI text, scripts, paths, or window titles, and provides `macos-harness telemetry disable` to turn telemetry off.

arc-code is a minimal harness for testing whether a general coding agent can learn unfamiliar ARC-AGI-3 games during play. It runs a headless coding agent with a shell, filesystem, and action command inside an internet-isolated sandbox; the harness provides no solver, planner, world model, grid tooling, fine-tuning, MCP servers, subagents, plugins, or hooks. The agent interacts with each game only through `act.py`, which executes actions and appends resulting boards and actions to a log. It uses the shell and approved file tools to parse that log, record findings, and build task-specific machinery such as parsers, rule models, simulators, and search procedures, which are discarded when the game ends. `run.py` launches and records one session per game, while the `rig/` directory supplies sandboxing, brokering, auditing, scoring, and export infrastructure. The repository includes the harness and six complete session records.