Reels often fail because creators optimize for generic short-form advice. This skill gives Reels-specific scripts, captions, trial decisions, and metric diagnostics.

Short-form videos often fail because the first seconds do not earn attention. This skill helps Claude, Codex, and Claude Code generate, critique, and refine hooks with visual, verbal, and text layers.

Plain subtitles can feel disconnected from a polished talking-head video. This skill builds readable rails, embedded hero captions, and themed motion treatments while preserving the original footage.

Complex coding changes often need more than one planning perspective. This skill runs several AI planner CLIs, compares their plans, and produces one consolidated implementation plan.

Browser automation agents are hard to wire together safely and consistently. This skill provides setup steps and a Playwright loop for Gemini Computer Use workflows.

Library APIs change faster than model knowledge. This skill queries Context7 so Claude, Codex, and Claude Code can reference current documentation during implementation.

Grasshopper workflows often require many repeated component, connection, and parameter steps. This skill provides scripts and guidance to turn MMD plans into executable Grasshopper MCP workflows.

Location work often requires separate searches, route calls, and map links. This skill combines AMap POI search, route planning, travel planning, and visualization workflows.

Building AMap web maps requires many small API choices across loaders, controls, overlays, and services. This skill gives Claude, Codex, and Claude Code structured references for producing JSAPI v2.0 map implementations.

Production ML work needs deployment, monitoring, and architecture decisions that are hard to standardize. This skill gives Claude, Codex, and Claude Code structured guidance for MLOps, LLM integration, and RAG planning.

Teams need repeatable deployment guidance without rebuilding DevOps checklists from scratch. This skill provides CI/CD, Terraform, deployment, and operations references for Claude, Codex, and Claude Code workflows.

Startup teams need secure AWS designs without overspending. This skill recommends serverless architectures, IaC options, cost controls, and deployment guidance.

Current research often needs fresh sources beyond the model context window. This skill sends a research prompt to Gemini CLI and asks for sourced Markdown answers.

Branch setup can waste time when each task needs a clean workspace. This skill creates a Git worktree and prepares local environment files.

GitHub login and repository authorization often need separate OAuth flows. This skill shows a two-connection Nango pattern for user identity, GitHub App access, and webhook processing.

Manual UI checks can miss visual regressions and take repeated effort. This skill guides browser automation with screenshots, element selection, and page inspection for verifiable web testing.

Video region cleanup is difficult when objects move across frames. This skill helps Claude, Codex, and Claude Code choose RunComfy routes for prompt-driven video inpainting.

Short AI video clips often stop before the story beat is complete. This skill guides Claude, Codex, or Claude Code to extend a user-provided video through RunComfy.

Short video generation often needs consistent characters, media references, and clear model settings. This skill guides Claude, Codex, and Claude Code through Seedance 2.0 Pro prompts and RunComfy CLI invocation.

Media model workflows often require manual setup, schema lookup, and polling. This skill gives Claude, Codex, and Claude Code practical RunComfy CLI guidance for repeatable terminal automation.