Video & Animation

MCP local - Code intelligence tools — with broad commands that allow agents in a huge repository to do almost anything: identify the right problem → find the right file/symbol → verify the right flow → measure the scope of impact → safely refactor → check API contracts → audit health, all quickly and accurately.

纸上算法 PAPER ALGORITHM · 试界TryWorld 官方 AI 技能合集：口播总入口 koubo、纸上算法出片 paper、AIHOT 选题 topics——一句话到成片的口播视频生产线

The practices that made software work didn't stop working. They stopped keeping up. PAAD brings them back at AI speed.

Agent skill that turns Claude Code / Codex into a motion-design studio for voiceover-driven explainer videos — word-level voiceover sync, 78 motion recipe cards, an anti-slideshow camera system, Remotion rendering.

The first vision plugin for DeepSeek Harness, and the vision bridge for every text-only coding agent. Paste an image, get structured JSON evidence (OCR, layout, semantics). | 全网最强 DeepSeek Harness 外挂视觉插件，为 DeepSeek、GLM 等纯文本模型外挂视觉能力，粘贴图片即得结构化 JSON 证据（OCR、版面、语义）。

Codex skill for evidence-backed short-video analysis, remix scripts, and storyboards.

Recently updated agent-skill-related GitHub repository: sgiffhorn/claude-skills-marketplace.

小滕房产分镜视频风格 Codex Skill 模板

生成带时间码字幕｜把本地音频或视频转成 SRT 字幕和 Remotion 时间轴的 Codex Skill

Claude skill: automate Meta App Review submissions (permissions, videos, submit)

A Codex skill for efficient, evidence-driven inspection of web pages, images, videos, and screen recordings.

Codex skill for converting local videos into moderate-contrast black/white depth-mask segments.

Your LinkedIn operating system inside an AI agent — eight open-source Agent Skills for Claude Code and Codex.

An autonomous development team for Claude Code: a clarifying intake agent, an orchestrating tech lead, scoped builder agents, and hook-enforced review gates.

Adaptive design direction and execution skill

Agent skill that stops LLMs from treating training-data conventions as facts about this repo.

Version control conventions for AI agents: commit grammar, branch flow, merge request and issue formats, scoped labels, semantic release bumps, and CI/CD.

Local-first podcast transcription and Show Notes archiving skill for Codex and OpenClaw

A bilingual, tool-agnostic workflow for turning knowledge stories into narrated illustrated videos. 中英双语知识故事视频制作工作流。

Governance layer for personal knowledge bases: a 4-tier quality gate, a single ledger with resumable batches, and anti-loss rules. Sits above conversion tools, not instead of them.

余生姜在真实业务中持续使用、测试和迭代的 AI Skills

Local OCR, color-statistics, pixel-scan, and metadata bridge as a cross-framework skill for text-only agent models; no external vision API. Algorithm-identical to dsh-pseudo-vision / pi-pseudo-vision.

A $50 million writing-style skill: restrained Chinese long-form in New Concept Writing Style, with a prose checker CLI

Subscribr CLI and AI agent skill for YouTube content creation