Say one sentence, get a multi-voice audiobook, kids' story or bedtime radio drama — voices, music, pictures, video. Claude Code skill. 有聲書自己做
Directory
Browse agent skills
Search by intent: design systems, testing, deployment, security, documentation, Azure, Supabase, React, or any workflow you want an agent to perform better.
Showing 1,621–1,650 of 125,967 skills
Page 55 of 4,199 · loading 30 at a time
一句话生成超写实人物三视图|Claude Code skill for Nano Banana Pro|One sentence in, photoreal character sheet out
System prompt assembly registry for the DeepSeek Harness
Session-projection seam: the merge-extensible projection type table, the provider contract, and the ctx.sessionProjections registry serving whole current values of log-derived per-session state
Conservative PDF paper extraction, agent translation validation and PDF reflow for DeepSeek Harness
DeepSeek Harness plugin: a Settings page for user-defined global system prompts, merged into every conversation
Placeholder: name reserved for RagScene, a docs-retrieval (RAG) product. No functionality yet.
Open-source agent skills for GRC practitioners: risk registers, policies, audit readiness, evidence, DPIAs and Cyber Essentials across ISO 27001, SOC 2, GDPR and ISO 42001.
An AI operating system for GRC practitioners, built on Claude Code. Clone it and run your compliance programme across ISO 27001, SOC 2, Cyber Essentials, GDPR and ISO 42001.
n8n nodes for Revid: create AI videos from prompts, scripts and articles, add captions, cut clips, generate images and publish to TikTok, YouTube, Instagram, X and Facebook.
Full-text search for hazejs models: database full-text (MySQL FULLTEXT, PostgreSQL tsvector, SQLite FTS5, LIKE), Meilisearch, Typesense and Elasticsearch / OpenSearch over fetch, with model index syncing
JavaScript bridge for the firstlightui/local-ocr NativePHP plugin: on-device document capture and OCR
A Claude Code skill that reverse-engineers any animation frame by frame, then proves the copy moves the same.
My complete Claude Code setup: plugins, skills, MCP servers, agents and commands, installed with one command.
SQL connection, catalog, query, and result workbench for the official DSH right Sidebar.
Send and change messages, manage channels, find users, add reactions, and upload files with a Slack bot, and start a workflow on a Slack event, such as a new message or a mention. Use it for team notifications and chat bots.
Send, edit, and react to messages, manage channels, members, and roles with a Discord bot, post with a webhook, and start a workflow on a slash command or a button click. Use it to build Discord bots, alerts, and moderation flows.
二宗卡片 · 苹果风小红书图文 / 公众号贴图:把文字排成 3:4 卡片,自动选版式、自动质检 | Apple-style Xiaohongshu (RedNote) & WeChat cards
Skill Claude : 45 effets vidéo en français
Neutral agent execution runtime kinds and implementations: persistent code sessions, one-shot sandboxes, and voice orchestration. Kept as a hard-wall package so the broad Agents UI/composition manifest cannot create platform/chat dependency cycles.
Tool descriptors, invocation contracts, and agent-facing primitives for Vibes applications.
AWS S3 and SecretsManager adapters plus AWS SDK OpenTelemetry instrumentation
DeepSeek Harness slash commands that submit conversation messages and run packaged research and development workflows through the Agent.
Unified LLM API with automatic model discovery and provider configuration
An OpenAI-compatible gateway in front of the free model stations (Cline, TRAE, ZCode, Qoder, MiniMax, Gemini, CodeBuddy, ...)
Official Unified TypeScript SDK & CLI for SuperApp Mini Apps: Core API, Bridge Transport, Mock Engine, and Browser Simulator
FLOP verified inference for Pi and Pi Durable, with model routing and replay-safe paid jobs
Official client SDK and viral meme metadata generator for Outstanding Tiny Desk AI (https://outstanding-tiny-desk.org) - Swap 1 to 5 portraits into the legendary Charlie Wilson NPR Tiny Desk closing scene with Seedance 2.5 video-to-video diffusion.
Browse the awesome-claude-video-skills catalog inside DSH and install a skill with one click.
Bilingual benchmark for LLM function-calling reliability with deterministic sandbox grading