See the writing-docs skill for details on writing documentation.
Guides stable API and interface design. Use when designing APIs, module boundaries, or any public interface. Use when creating REST or…
Run UnrealCV closed-loop test workflow (build + launch + test)
Use when working with react-native-skills tasks or workflows
>
Build Claude agents using TypeScript with the @anthropic-ai/claude-agent-sdk.
Debug Streamlit frontend and backend changes using make debug with hot-reload.
Manages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to…
Master Godot 4 GDScript patterns including signals, scenes, state machines, and optimization.
Create new Claude Code Skills with proper YAML frontmatter, progressive disclosure structure, and complete directory organization.
Create and manage Claude Code skills following Anthropic best practices.
SwiftUI 前端设计 skill — anti AI-slop rules, design direction advisor, brand asset protocol, and five-dimension review. Works with Claude Code, Cursor, Codex, and OpenCode.
Build 3D web apps with Three.js (WebGL/WebGPU).
Use when you have lint errors, formatting issues, or before committing code to ensure it passes CI.
Generate API documentation from source code with JSDoc and OpenAPI support
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use…
Clean Architecture patterns for Android and Kotlin Multiplatform projects — module structure, dependency rules, UseCases, Repositories…
Automate Appcircle tasks via Rube MCP (Composio).
Use this skill when inspecting Blender characters, rigs, poses, animation retargeting, ground contact, facing direction, or…
Creates GitHub pull requests with properly formatted titles that pass the check-pr-title CI validation.
Create, install, or update skills in the workspace.
Developer Tools & Repos agent skills on mcprush govern how an agent reviews, debugs and ships code: what a review comment is allowed to say, when a bug counts as reproduced, what a flaky test costs before it may be quarantined. The kinds are workflow, expertise, output format, voice & style and guardrail, and the one a skill belongs to is the fastest way to guess what it will do to your prompt.
What to look at before installing one is what it costs in context and what it takes away: a guardrail that refuses an action and a voice that rewrites a sentence are both skills, and only one of them will stop the agent doing something. Every skill here says which kind it is, what it costs in context, what it needs beside it to work and the licence it is published under, next to the publisher it came from and whether that publisher has been verified. The filters on the left narrow it further and the sort at the top decides what comes first.
A skill is not connected to anything: there is no endpoint to authorise and nothing routed through the gateway. Installing one drops a folder of instructions into the client’s own skill folder, and uninstalling it is deleting that folder. The exact line for each client is on the skill’s own page, because it differs by client rather than by skill.