Meta Launches Muse Code, a Terminal AI Agent Built for Large Code Bases

Meta released a new AI coding agent called Muse Code on Wednesday, August 6, 2026, targeting programmers who need help managing complex tasks across large software repositories. The tool is currently available in beta and can be installed with a single command.

According to Meta CEO Mark Zuckerberg, Muse Code can handle “complete software engineering tasks across large repos,” including “planning changes, writing code, validating the results.” The agent is powered by Meta’s existing coding model, Muse Spark.

A key feature of Muse Code is its ability to spin up parallel sub-agents when a task is large enough. “When a job is big enough, it fans out to separate sub-agents working in parallel in isolated worktrees,” Zuckerberg explained. “Your working copy is never touched.” He noted that in testing, the system built six game features simultaneously without conflicts.

Meta positions the tool as a cost-competitive alternative to similar offerings from rivals. “We think that for a lot of workflows and a lot of use cases, this can be an incredibly good option, especially from a cost perspective,” said Alexandr Wang, Meta’s AI chief and head of Meta Superintelligence Labs, in comments to the Wall Street Journal. Muse Code enters a market that includes OpenAI’s Codex and Anthropic’s Claude Code.

The launch reflects Meta’s broader push to expand its presence in AI beyond its core advertising business. In June 2026, the company entered the enterprise AI market with an agent aimed at customer service and support. Muse Code represents a further step in that direction, this time targeting software development workflows.

Source: TechCrunch

This article was generated by AI and cites original sources.
Scroll to Top