Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro). Use when you need to: (1) generate images from text descriptions, (2) edit existing images with AI, (3) compose multiple images into one scene. Supports 1K/2K/4K resolutions and up to 14 input images for composition.
This skill is designed to rewrite user prompts to align with the expert-level aesthetic standards. It transforms simple descriptions into multi-dimensional, professional-grade artistic instructions that maximize scores across all fine-grained aesthetic attributes.
This skill should be triggered when the user's request implies a need for imagination, artistic flair, conceptual depth, or unconventional visual storytelling.
This skill should be triggered when the user's request involves multiple objects, complex scene arrangements, or specific physical relationships between elements.
A skill for generating images with MiniMax from text descriptions or from other images. It supports common image shapes such as square, widescreen, portrait, and standard photo proportions.
A video-generation tool for creating short MP4 videos from text, images, video frames, or a subject reference. It uses the MiniMax Hailuo model and saves generated videos locally.
Work on the adopt module — the ordered pipeline that adopts Claude Code into a GitHub repo (toolchain, clone, branch, trust, init, conform, enforcer, verify, push, PR), its step contract, CLI flags and credential masking. Use when adding or changing an adoption step, debugging a run, or when the user says "adopt"…
Write, test and wire a claude-code-enforcer rule — the custom maven-enforcer rules that fail the build when CLAUDE.md, AGENTS.md, README.md or the .claude configuration is malformed. Use when adding or changing a rule, when configuring severity/reportFile/baselineFile, or when the user says "enforcer rule", "the doc…
Systematic Java code review for the tools repo — leads with the ArchUnit-enforced rules the build fails on, then the five defect shapes this repository actually ships fixes for, then null safety, exceptions, concurrency, and performance. Use when the user says "review code", "check this PR", "code review", or before…
Use when the user wants to run ComfyUI image/video generation on a rented remote GPU instead of paid API subscriptions: rent a GPU box, bootstrap the stack, connect the agent over SSH, drive workflows via the ComfyUI API, and deliver results to Telegram.
Turn a folder of static HTML into a live commenting surface and act as the agent that responds. Inject a client library, start a local server, watch an on-disk inbox; when the user highlights text, picks an element, drags a region, or asks a question, read it and edit the HTML in response. If the user has no HTML yet…