Meta AI has released Muse Code (in beta), a terminal coding agent powered by its new Muse Spark 1.2 model, positioning the pair as its next step toward the frontier. Muse Code targets complex software engineering across large repositories by planning changes, writing code, and validating results. It uses async background agents that remain active for the whole session, avoiding redundant information gathering. A local append-only event log records every model call, tool run, approval, and edit, which Meta calls replay-exact and restart-safe, allowing the agent to resume after a crash. Muse Spark 1.2 was co-trained with the harness itself. Meta published a kernel-optimization case study running 1,000+ tool calls over as long as 24 hours. Muse Code ships in beta for macOS and Linux via curl -fsSL https://dev.meta.ai/install.sh | bash. Muse Spark 1.2 is available in Muse Code and the Meta Model API with expanded global access. The launch post does not mention downloadable weights, so treat this as a hosted dependency. The agent includes three default skills: /plan, /grill, and /goal. Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1, with gains in code generation, complex debugging, codebase understanding, and end-to-end developer workflows, achieved by scaling up training compute and expanding environment diversity.