Remove backgrounds from images using local AI (rembg). Use when removing backgrounds from character art, mascot images, photos, or any image that needs a transparent background.
Fetch CURRENT library/framework/API/CLI documentation via Context7 instead of relying on training data. Use whenever the user mentions a specific library, framework, SDK, API, or cloud service, asks a setup/configuration question, plans a version migration, or is debugging library-specific behavior. Triggers on common web frameworks, APIs, and cloud services — even well-known ones, since training data may be stale.
Test and deploy changes safely. Discovers deploy targets, runs fail-stop gates before going live, optionally shadow-deploys and swaps, then runs report-only post-deploy checks. This is a TEMPLATE — customize the commands and checks for your specific deployment pipeline.
Review uncommitted code changes using parallel Claude sub-agents (Bug Hunter, Rules Auditor, optional Architect). The invoking agent triages the diff by file path into impacted modules and risk surfaces, then spawns reviewers scaled to the change. Each reviewer self-primes via /prime, verifies any API/library claim via Context7 (mandatory — unverified claims are auto-discarded), and reports an intent verdict against progress.md before its findings. Catches bugs, security issues, CLAUDE.md compliance, and test-coverage gaps. Skip for trivial typos/formatting. Use after substantive implementation work, or when the Stop hook requests it. Also invocable manually with /review-work.
Get a second opinion from Google's Gemini Pro via the locally installed Gemini CLI (defaults to gemini-3.1-pro-preview; override with the CLAUDE_SECOND_OPINION_MODEL env var). This is the explicit / fallback engine — invoke ONLY when the user explicitly says "ask Gemini", "Gemini's take", "what does Gemini think", or "ask both" (run alongside the default engine and synthesize). Do NOT trigger on a generic "second opinion", "another perspective", or "cross-check this" — those route to the default `second-opinion` skill. Reserve this for when the user specifically wants Gemini, or as a fallback when the default engine is unavailable. Reports unavailability rather than falling back to a weaker model.
Get a second opinion from OpenAI's Codex CLI running locally. Use this skill when in Plan Mode for large or critical tasks, when stuck on a debugging dead end, when facing architecture trade-offs, for subtle edge cases in code review, or any situation where an independent perspective would add value. Also use when the user explicitly asks for a "second opinion", "another perspective", "cross-check this", "ask Codex", or "ask GPT". This is the DEFAULT second-opinion engine — Codex runs a completely different model architecture from Claude, so it catches blind spots Claude shares with itself. For a Gemini second opinion specifically, the user says "ask Gemini" (separate second-opinion-gemini skill); "ask both" runs this skill and that one, then synthesizes.
Update project documentation after code changes. Maintains the 4 core ai-context files (spec, project-structure, progress, deployment-infrastructure) and CLAUDE.md. Use after completing features, refactors, or any changes that affect project structure, capabilities, or status. Also creates initial documentation if files don't exist yet.
Edit images with precision — crop, resize, mirror, rotate, trim, and reframe. Use this skill whenever the user asks to crop, resize, trim, mirror, flip, rotate, reframe, or otherwise manipulate an image. Also use for creating square crops, portraits/headshots from full-body images, icon sizes, or any image transformation. Even if the request sounds simple, this skill prevents common pitfalls and ensures correct results on the first try.