This skill should be used when the user asks to "catch me up", "what's on this branch", "summarize this branch", "what changed since main", or wants a structured summary of commits, changed files, and current state of the working branch compared to main/master.
This skill should be used when the user asks to 'debug this', 'fix this bug', 'why is this broken', 'fix CI', 'pipeline failure', 'fix failing tests', 'investigate this error', 'not working', or when troubleshooting unexpected behavior, exceptions, or runtime errors.
This skill should be used when the user asks to "fix this bug", reports a Sentry issue, provides an error report, stack trace, screenshot, or production error and wants an end-to-end fix from diagnosis through a reproduction test to validation.
This skill should be used when the user asks to 'write tests', 'add tests', 'create test cases', 'cover this with tests', 'unit tests', 'integration tests', 'test coverage', 'test this function', 'fix failing tests', or when implementing tests as part of a feature or bug fix task.
This skill should be used when the user asks to 'humanize text', 'remove AI writing patterns', 'make this sound more human', 'fix AI writing', 'edit for natural voice', or when about to draft user-facing prose: a pull request description, merge request description, changelog entry, commit message body, README section, blog post, design doc, release note, or any documentation paragraph. Trigger proactively before submitting written text, not only on explicit 'humanize' requests. Detects and removes inflated language, promotional tone, em dashes, rule of three, AI vocabulary, passive voice, negative parallelisms, and filler phrases per Wikipedia's 'Signs of AI writing' guide.
This skill should be used when the user asks to 'address review comments', 'respond to review', 'reply to reviews', 'fix review feedback', 'handle PR feedback', 'address code review', 'review suggestions', 'mark as fixed', 'resolve review comments', 'reply to copilot', 'reply to sourcery', or needs to apply changes, reply to, or resolve PR/MR review comments.