Automate end-to-end code implementation and testing with persistent coding and testing subagents, iterating until tests pass and committing results locally.
Initial release of workflow-runner: automate end-to-end coding and testing workflows using persistent subagents. - Parses requirements, splits tasks between coding and testing subagents, and loops until tests pass. - Subagents operate in isolated 128k contexts and persist for up to 24 hours (configurable). - Produces all artifacts in results/ and commits code locally (no remote push). - Supports configuration for TTL, max retries, commit behavior, and results path. - Triggered by natural language requests or explicit `run workflow-runner: <spec>` commands. - Includes example workflow and orchestrator/helper scripts.