From kanban to harness: when the tracking tool becomes the orchestrator
When I shipped KittyClaw two weeks ago, the tool did one thing: serve as a board. The Claude agents ran alongside - first by hand, then via a dispatcher.mjs : a Node script polling KittyClaw's API, triggering the right agent based on who was assigned to which ticket. The dispatcher worked great. It orchestrated Aekan's 13 agents for weeks. But it was an external process : one more node…
KittyClaw, a kanban orchestrator, was initially used solely as a board to facilitate AI agent tasks. The previous setup involved two separate processes - KittyClaw as the board and dispatcher.mjs as an external Node script responsible for orchestrating Aekan's 13 agents based on ticket assignments. While effective, this setup required additional infrastructure, configuration, and manual work to maintain across projects.
As KittyClaw evolved, the dispatcher process was integrated directly into the board. The automations file, defined within the project workspace, now outlines triggers, actions, and their respective conditions and parameters. This streamlined approach reduces redundancy, improves efficiency, and centralizes the orchestration within KittyClaw itself.
While the shift from an external dispatcher to an internal one may appear minor in terms of code changes, it significantly alters the tool's purpose and the user's interaction with it. The new workflow is simpler: users create tickets, and KittyClaw autonomously assigns them to agents, launches AI agents, and monitors progress, all within the same interface.
This change emphasizes the importance of user-defined automation and monitoring, allowing users to focus on higher-level tasks such as deciding what tasks to automate, reviewing outputs, and refining the system.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
