Headless browser automation CLI optimized for AI agents. Uses snapshot + refs system for 93% less context overhead vs Playwright. Purpose-built for web testing, form automation, screenshots, and data extraction.
Manage a gitflow branching workflow — starting and finishing feature, release, and hotfix branches; cutting versioned releases with changelog generation; coordinating emergency hotfixes directly to production; and keeping long-lived branches in sync. Activates when users mention gitflow, feature/release/hotfix branches, cutting a release, branching strategy, promoting an integration branch to production, tagging a version, or rolling back a live release. Also reaches for it when the user says "ship it", "promote dev to main", or "we need to hotfix prod" without naming gitflow. Skip for general git mechanics (commit messages, merge conflicts, interactive rebase, git education) or CI-failure debugging unrelated to a release.
Fixes a bug through test-driven debugging — reproduces it with a failing test, locates the root cause with evidence (file:line), then applies the smallest fix that resolves it without refactoring unrelated code. Use when the user wants to fix a bug, debug an issue, resolve an error, or investigate a failing test. Not for building new functionality (use implement-feature) or restructuring working code with no bug involved (use refactor).
Implements a new feature end-to-end as a senior staff engineer would — discovers project conventions, researches current best practices, drafts a plan for approval, then builds it with parallel subagents that reuse existing code, skip speculative abstractions, and verify with tests before completion. Use when the user wants to implement, build, add, or ship new functionality (a feature, endpoint, component, module, or integration) — not for fixing an existing bug (use fix-bug) or restructuring code that already works (use refactor).
Restructures existing code without changing its behavior — maps callers and test coverage, adds characterization tests where coverage is thin, then applies the change in small steps verified against the full test suite after each one. Use when the user wants to refactor, extract a method or class, simplify logic, reduce duplication, improve naming, restructure modules, or pay down technical debt in code that already works. Not for adding new functionality (use implement-feature) or fixing broken behavior (use fix-bug).
Runs a comprehensive multi-agent code review of a PR, commit, or the whole codebase across six dimensions (correctness, performance, code style, test coverage, error handling, and simplicity/over-engineering) and returns a severity-ranked report with file:line findings and fix suggestions. Use when the user wants a thorough code review, asks to review a PR or diff, or wants over-engineered code flagged for simplification. Analysis only, identifying issues without modifying code, committing, or running tests. Not for a security-focused audit (use review-security) or a visual/UX design critique (use review-design).
Generates a test suite by analyzing coverage gaps, prioritizing critical and untested code paths, then writing tests in parallel that match the project's existing patterns. Use when the user wants to write tests, add test coverage, generate test cases, improve testing, or analyze coverage gaps. Supports pytest, vitest, jest, and all major test frameworks. Not for debugging a specific failing test (use fix-bug).
Create new agent skills with specification-driven generation, live documentation fetching, and interactive planning. Use this skill whenever the user wants to create a new skill, slash command, or agent capability — even if they say 'make a command' or 'turn this into a reusable workflow'.