superpowers

Systematically address GitHub PR review comments one at a time, with interactive guidance and resumable progress

Full PR lifecycle — local code review and test loop, PR creation, Copilot review loop, CI verification, and merge conflict resolution

Walk through a pull request interactively, explaining code changes to help you understand how everything works. Supports GitHub and Azure DevOps.

Use this agent when a major project step has been completed and needs to be reviewed against the original plan and coding standards. Examples: <example>Context: The user is creating a code-review agent that should be called after a logical chunk of code is written. user: "I've finished implementing the user authentication system as outlined in step 3 of our plan" assistant: "Great work! Now let me use the code-reviewer agent to review the implementation against our plan and coding standards" <commentary>Since a major project step has been completed, use the code-reviewer agent to validate the work against the plan and identify any issues.</commentary></example> <example>Context: User has completed a significant feature implementation. user: "The API endpoints for the task management system are now complete - that covers step 2 from our architecture document" assistant: "Excellent! Let me have the code-reviewer agent examine this implementation to ensure it aligns with our plan and follows best practices" <commentary>A numbered step from the planning document has been completed, so the code-reviewer agent should review the work.</commentary></example>

Expert code refactoring that strictly preserves existing behavior. Use when asked to refactor code, improve code structure, reduce duplication, extract methods/classes/modules, simplify complex conditionals, improve naming, apply design patterns, reduce complexity, clean up code, make code more maintainable/readable/testable, decompose large functions/classes, or find cross-file patterns and commonalities to consolidate. Language-agnostic. Triggers on requests like "refactor this", "clean up this code", "this function is too long", "reduce duplication", "simplify this logic", "make this more maintainable", "make this more object-oriented", "apply design patterns", "find common patterns", or "what can be consolidated across files".

Use when making architectural decisions that need documentation, after brainstorming explores trade-offs, or when code review reveals undocumented decisions - captures decision context, alternatives, and rationale in standardized Architecture Decision Record format at the moment decisions are made

Monitor an Azure DevOps (ADO) pipeline run to completion — the whole run or a single stage/job — using a blocking poller that estimates expected duration from past runs, adapts its polling cadence, holds control until a final outcome is reached, and returns a structured JSON result. Use this whenever the user wants to watch, wait on, track, poll, or be told the outcome of an ADO/Azure Pipelines build or release, asks 'tell me when this pipeline/job finishes', 'is this build done yet', 'wait for buildId NNN', or pastes a dev.azure.com/_build/results URL and wants its result. Prefer this over ad-hoc repeated status checks by the agent.

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

Use when determining what type of Claude Code extension (command, skill, or agent) a prompt or description should be classified as. This skill should be used when users ask "should this be a command, skill, or agent", when designing new extensions, or when there is ambiguity about the correct extension type for a given use case.