Repo Feature Planner
Explores your repository and returns a commit sized plan, the risks and a definition of done before it writes code.
You fill in describe the feature you want…
A Claude Code prompt that surveys every use of a package, lists the breaking changes that touch you and upgrades it one commit at a time.
Surveys every use of a package, lists the breaking changes that touch you and upgrades it one commit at a time.
I want you to act as an engineer upgrading a dependency across a major version. I will name the package and the version to move to. Survey first. Find every place the package is used, including use through our own wrappers, configuration and build scripts, and report the count and the files. Read the changelog for the versions between the one installed and the target, and list only the breaking changes that touch code in this repository, each with the file affected. Do not change anything yet. Then tell me whether to do this in one step or through intermediate versions, and why. Give the order of work as commits: the mechanical replacements first, the behavioural changes after, and the test for each behavioural change written before it. When I approve, carry it out one commit at a time, running the tests after every commit and stopping at the first failure. Do not upgrade other packages at the same time, do not touch the lock file beyond what this change needs, and do not silence a deprecation warning by hiding it. Finish with the diff, the test results, every deprecation warning still open, and the parts of the upgrade the existing tests cannot verify, with what I should check by hand. My first request is: "Upgrade [package] from [current version] to [target version]."
Explores your repository and returns a commit sized plan, the risks and a definition of done before it writes code.
You fill in describe the feature you want…
Writes a failing test first, explains the real cause, then makes the smallest fix and runs the whole suite.
You fill in describe the bug, how to…
Refactors behind passing tests, writing characterisation tests first and stopping the moment one of them fails.
You fill in file or modulethe reason
Reviews a branch or pull request for defects, risks and missing tests, with a verdict and no invented findings.
You fill in branch name, pull request…
Reproduces a failing build locally, names the real cause and fixes it without skipping or deleting a single test.
You fill in paste the failing log herebranch name
Reads your repository and writes its CLAUDE.md: verified commands, the layout and the conventions the code follows.
You fill in how it is deployedthe command I actually run…what an agent should know
One email a week: the best new prompts and one short guide. Unsubscribe any time. Privacy policy