Just shooting from the hip, I don’t see this scaling super well. As the number of tickets grows you’ll need some way to efficiently track WIP, filter tickets by status, etc. Jamming it all into LLM context won’t do, and relying on the agent harness to automatically sort it out will be unreliable.

So you’ll eventually want a more formal way to track and query everything, and to route the coding agents through that API to ensure consistency and reliability. At which point you’ve basically just reinvented the ticketing system that you originally sought to eliminate. Only it might still be flaky because every source control branch introduces a secondary source of truth for ticket status.