← ClaudeAtlas

da-fix-planlisted

Turn a review report into an ordered fix plan on disk. Use after a code review or PR feedback, when there are more findings than you want to act on. Decides what NOT to fix and why, orders what remains by irreversibility and dependency, and writes the plan to a file. Read-only until the plan is agreed.
bwkw/dotagents · ★ 0 · Code & Development · score 76
Install: claude install-skill bwkw/dotagents
# /da-fix-plan — decide what not to fix, then order the rest A review produces findings. This turns them into a plan, and **its primary job is subtraction.** > A reviewer prompted to find gaps **will report some even when the work is sound**, because that is > what it was asked to do. Chasing every finding leads to over-engineering: extra abstraction layers, > defensive code, and tests for cases that cannot happen. So the failure this skill exists to prevent is not "a finding got missed". It is **acting on all of them**. A plan that accepts every finding has not triaged; it has transcribed. **Read-only until the plan is agreed.** Produce the plan, show it, and stop. Fixing happens after. ## Preconditions | Condition | If unmet | |---|---| | A review report exists — in this conversation, at a path, or on a PR | **Stop.** Ask which review. Never plan against remembered findings. | | You can tell what the change was *supposed* to do — a spec, a plan, a PR description, or the user saying so | **Ask.** Without it you cannot separate "the code is wrong" from "the reviewer wanted something else", and that distinction is most of the work here. | | The findings carry locations | Continue, but mark any finding you cannot locate as **unactionable** rather than guessing at one | ## Position in the workflow | Upstream | This skill | Downstream | |---|---|---| | `/da-review-all`, `/code-review`, `/find-bugs`, or a human review | `/da-fix-plan` | `/executing-plans` on the accepted s