Harness
Agentic coding tools you run in your terminal — the agent loop itself.
Block's open-source workspace where humans and AI agents share rooms — Nostr identities, ACP harness, agent-first CLI, and self-hosted relay.
Agentic coding tool that lives in your terminal, understands your codebase, and executes routine tasks, explains code, and handles git workflows via natural language.
Lightweight coding agent from OpenAI that runs locally in your terminal, usable with a ChatGPT plan or API key.
MicroVM-isolated sandboxes for running AI agents — instant, concurrent, and secure, with a template store and web console.
A filesystem-first framework for durable AI agents — instructions, tools, skills, channels, and schedules live as conventional files you can inspect, extend, and operate.
Open-source AI agent that brings Gemini directly into your terminal, using a ReAct loop with built-in tools and MCP server support.
Brings the Copilot coding agent to the terminal, letting you build, debug, and understand code via natural language.
SpaceXAI's terminal coding agent harness and TUI — fullscreen, mouse-interactive, extensible, with headless and ACP modes.
The agent that grows with you — a cross-platform AI agent that installs with one command and bundles its own runtime.
Command-line companion and TUI for Jules, Google's asynchronous cloud coding agent, to create and manage agent sessions that write tests, fix bugs, and open pull requests.
A single-binary AI coding-agent CLI for the terminal — reads and edits code, runs commands, and speaks the Agent Client Protocol to drive Zed and JetBrains.
AWS's agentic command-line tool for building software through natural-language chat in your terminal, the successor to the Amazon Q Developer CLI.
Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1
The meta-harness for AI agents — scaffold your own focused, branded agent harness with its own npx CLI, MCP server, memory, and learning loop.
A terminal-native AI coding agent with persistent cross-session project memory and built-in multi-agent switching; a free zero-config channel to start.
A ~100-line, dependency-light SWE agent from the SWE-bench team — a clean reference harness for coding-agent research.
A multi-model coding-agent harness that installs into OpenCode and the OpenAI Codex CLI.
A terminal coding agent with the IDE wired in — LSP, DAP, subagents, browser, and 40+ providers.
A meta-harness that orchestrates Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, with policy enforcement and sandboxing.
Your own personal AI assistant for any OS and platform — installs via npm and runs as an always-on gateway daemon.
An open-source, model-agnostic coding agent for your terminal — bring your own model across 75+ providers and self-host it.
Open-source sandbox runtime for executing AI-generated code, with Docker and Kubernetes backends, SDKs, a CLI, and an MCP server.
An agent development environment (ADE) for running a fleet of parallel coding agents in isolated worktrees, on desktop and mobile.
A self-extensible coding-agent harness and AI agent toolkit: an interactive coding agent CLI with a unified multi-provider LLM API, TUI, and web UI libraries.