Claude Code orchestrators, ranked by freshness

The frameworks people put around a coding agent divide into two kinds, and mixing them up is the most common mistake in this market. LangChain, CrewAI and AutoGen are runtimes: you import them and they run the agents. Spec Kit and BMAD are processes: they produce the specification and the decisions, and the coding agent you already have does the running. If your problem is that the agent builds the wrong thing, a runtime will not help you, and vice versa. Within the runtimes, AutoGen is out for new projects on its own say-so. "maintenance mode … will not receive new features or enhancements". Leaving CrewAI for role-based teams and LangChain for integration breadth. Within the processes, Spec Kit is agent-agnostic and version-pinned and comes from GitHub; BMAD is Node-installed, sizes its process to the change, and is the only one of the five to state it has no paid tier. The number that decides between any of them is not in the table: four of the five say nothing whatsoever about whether they are still maintained, which is why this page ranks on freshness measured nightly rather than on anything a README claims.

Every list of these you can find today was written once and never re-measured, which is the whole problem, because the star counts kept rising after the commits stopped.

The ranking above is not an opinion about which orchestrator is best designed. It is a reading of which ones somebody is still working on, taken last night, with the signals behind each row one press away on the board.