Harnesses
Coding-agent tools with instruction-file guidance, context handling, permissions, and cautions. Harness behavior stays separate from model behavior.
7 / 7
- AiderAider · activeaiderAider is a terminal-native AI pair-programming harness. This entry covers its repository context, conventions, edit, and verification workflows, not the behavior of the model selected for a session.No declared filesCurrent · verified Sep 17, 2026
- Claude CodeAnthropic · activeclaude-codeClaude Code is Anthropic's coding-agent harness. These notes describe the tool environment, not the underlying Claude model.CLAUDE.mdCurrent · verified Sep 7, 2026
- CodexOpenAI · activecodexCodex is OpenAI's coding-agent harness. This entry covers harness behavior rather than behavior of any underlying model.AGENTS.mdCurrent · verified Sep 7, 2026
- CursorCursor · activecursorCursor is an editor and CLI coding-agent harness. These notes cover its rules, context, and permission behavior rather than the underlying model.AGENTS.mdCLAUDE.md.cursorrulesCurrent · verified Sep 17, 2026
- Gemini CLIGoogle · activegemini-cliGemini CLI is Google's terminal coding-agent harness. This entry describes its context-file behavior, not behavior inherent to the underlying Gemini model.GEMINI.mdCurrent · verified Sep 8, 2026
- GitHub Copilot CLIGitHub · activegithub-copilot-cliGitHub Copilot CLI is a terminal-native coding-agent harness. This entry covers its instruction discovery, session controls, and permissions, not the behavior of whichever model it runs.AGENTS.mdCLAUDE.mdGEMINI.mdCurrent · verified Sep 17, 2026
- OpenCodeOpenCode · activeopencodeOpenCode is a terminal-native coding-agent harness. This entry covers its instruction loading, agent modes, and permission boundaries, not the behavior of the model selected for a session.AGENTS.mdCurrent · verified Sep 17, 2026