No comments yet.
How it works
1. Describe the goal (Task Query) – In Phase mode, type a concise description of what you want to build or change. Example: “Add rate‑limit middleware and expose a /metrics endpoint.” Traycer treats this as the parent task. 2. Clarify intent (AI follow‑up) – Traycer may ask one or two quick questions (constraints, coding style). Answer them so the scope is crystal‑clear. 3. Auto‑generate the Phase Board – Traycer breaks the task into a sequential list of PR‑sized phases you can reorder, edit, or delete. 4. Open a phase & generate its plan – get a detailed file‑level plan: which files, functions, symbols, and tests will be touched. 5. Handoff to your coding agent – Hit Execute to send that plan straight to Cursor, Claude Code, or any agent you prefer. 6. Verify the diff – When your agent finishes, Traycer compares the diff to the plan and checks compatibility with upcoming phases, flagging any mismatches. 7. Review & commit (or tweak) – Approve and commit the phase, or adjust the plan and rerun. Then move on to the next phase.
Why it helps?
* True PR checkpoints – every phase is small enough to reason about and ship. * No runaway prompts – only the active phase is in context, so tokens stay low and results stay focused. * Tool-agnostic – Traycer plans and verifies; your coding agent writes code. * Fast course-correction – if something feels off, just edit that phase and re-run.
Try it out & share feedback
Install the Traycer extension (https://traycer.ai/installation), create a new task, and the Phase Board will appear. Add a few phases, run one through, and see how the PR‑sized checkpoints feel in practice. If you have suggestions that could make the flow smoother, drop them in the comments - every bit of feedback helps.
