Meta Launches Muse Code, an AI Coding Agent for Large Repositories

Meta has released Muse Code, a terminal-based AI coding agent currently in beta, designed to assist programmers with complex tasks across large software code bases. The tool can plan changes, write code, and validate results, as stated by CEO Mark Zuckerberg in a social media post on Wednesday. Muse Code is powered by Meta's earlier coding model, Muse Spark, and can be installed with a single command. It handles large projects by spawning sub-agents that work in parallel in isolated worktrees, ensuring the user's working copy remains untouched. Zuckerberg noted that in testing, it successfully built six features for a game simultaneously without collisions. This launch positions Meta more competitively and affordably against AI labs like OpenAI's Codex and Anthropic's Claude Code. Alexandr Wang, Meta's AI chief leading Meta Superintelligence Labs, told the Wall Street Journal that Muse Code could be an incredibly good option for many workflows, especially from a cost perspective. Meta has been increasing its AI investments, and in June, it expanded into the enterprise AI market with a customer service agent, moving beyond its core advertising focus.