Enables git-branch-style A/B experimentation for long-running tasks, allowing users to create, compare, promote, discard, or roll back parallel conversation branches via MCP tools.
Enables long-running coding or documentation tasks to save incremental checkpoints, resume after interruptions, restore any previous step, and hand off work across sessions or machines using only the Python standard library.
Enables deterministic, resumable workflow automation through DAG step graphs, boolean gate branching, automatic parallel execution, and in-step agent loops with sandboxed file and command tools.