Agent skills for Claude Desktop

ChatGPT bridge workflow skill

Read-only budget scout before implementing.

Use before claiming work is complete, fixed, or passing.

Framework-specific patterns for integrating CometChat React UI Kit v6 into React Router projects (v6 library mode and v7 framework mode).

Default workflow for generating, editing, transforming, or using raster images as references with the configured personal…

Generates traditional Chinese ancestral worship envelope (冥包/袱包) kinship titles and inscription text based on family relationships.

Co-creates researched, platform-adapted social content and standalone HTML previews for Xiaohongshu, Douyin, Weibo, Bilibili, and Kuaishou.

Picks which Cursor model a task should run on for a Pro Plus account (Composer 2.5 Fast, Grok 4.6, Auto Cost, GPT-5.6 Sol/Terra/Luna…

Demonstrate Insufficient Flow Control (CICD-SEC-1) during an authorized assessment — pushing code, config, or artifacts through the…

Detect and prevent CI/CD secret exposure (CICD-SEC-6).

Classify CI failures — distinguish clear regressions from infra flakes and security-test failures.

Create cinematic scroll-driven landing pages with Lenis smooth scrolling, GSAP ScrollTrigger, scroll-linked progression, staggered text…

Use this skill whenever the user's instruction is too generic or unspecific to act on directly, whether or not a file is attached.

Make user-facing narrative text clear, precise, concise, and useful.

Use Claude Code as an independent `claude -p` subagent when the user explicitly asks for Claude, wants a second-agent opinion from Claude…

Use when Codex should ask Claude CLI for an adversarial challenge review that pressure-tests implementation direction, design choices…

Build Claude agents using TypeScript with the @anthropic-ai/claude-agent-sdk.

A research-journal aesthetic printed on warm stone — authoritative, editorial, almost achromatic.

Performance monitor for Claude Code and local system.

Use when Codex should ask Claude CLI for a rescue-style investigation, second opinion, debugging pass, or explicitly requested follow-up…

Analyze a repository to generate recommended Claude Code settings.json permissions.