Executors & Harnesses | exo docs

The built-in executor runtimes, from basic to Codex, Claude Code, and Cursor.

Executors & Harnesses | exo docs Skip to content Menu Return to top Executors & Harnesses ​ A harness is an executor running on the exoharness. The exo CLI selects one with --harness : Harness What it is basic Built-in Rust executor: straightforward prompt → model → tools loop rlm Recursive-language-model experiment typescript Runs a TypeScript harness module that owns the turn loop, while Rust owns durable state codex Backs OpenAI Codex with durable exoharness sessions claude-code Backs Claude Code with durable exoharness sessions cursor Backs the Cursor SDK with durable exoharness sessions <

saved by

• Maanav Khaitan
• SUJASH AGRAWAL

related reading

• Data Model | exo docsexoharness.ai
• exoharnessexoharness.ai
• Game & Third-Party Tool Integration | exo docsexoharness.ai
• What is an Agent Harness?rubriclabs.com
• Welcome to Learn Harness Engineeringwalkinglabs.github.io
• Best practices for Claude Code - Claude Code Docsanthropic.com
• Towards self-driving codebases · Cursorcursor.com
• A Guide to Claude Code 2.0 and getting better at using coding agents – sankalp's blogsankalp.bearblog.dev
• Effective harnesses for long-running agents \ Anthropicanthropic.com
• GitHub - shareAI-lab/learn-claude-code: Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1github.com
• Continually improving our agent harness · Cursorcursor.com
• GitHub - affaan-m/ECC: The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.github.com