Writing & Editing

Detect and rewrite AI writing patterns to make Traditional Chinese text sound natural and human for clearer, more authentic communication.

Recently updated agent/AI repository: mehmetbaykar/claude-code-docs-skill

Agent skill that produces sharp, specific, human-sounding text. Forces audience-first thinking, frontloaded structure, and ruthless self-editing. The skill itself is written in the style it teaches -- every token in context biases the model's output distribution. 119 lines, ~900 tokens.

Bilingual (Korean/English) legal writing agent for Claude Code — drafts and revises non-contract legal documents with jurisdiction-appropriate conventions

Measure twice, cut once. A CLI tool for managing spec-writing workflows with AI agents.

Recently updated agent/AI repository: Manutsanan/claude-skill-copy

An agent SKILL that turn project spec files into a forest of feature trees, cluster features into dependency islands, render Obsidian-friendly Markdown, and track per-feature progress with checkbox round-trips.

AutoGen multi AI agent blog post writing using reflection

Agent skills for scientific paper writing

AI novel writing with multi-agent simulation: MiroFish backbone x AutoStory editorial board x nuwa-skill style distillation

A MCP extension built on top of rtfm-ai that it acts as a gatekeeper by taking the agent's task, queries the underlying RTFM index, filters out background noise, and packs only the essential and supporting source chunks into a tight contextualized prompt.

AI agent for IBM DOORS Next (DNG), EWM, and ETM — MCP server with 20 tools for reading, writing, and managing requirements, work items, and test cases across the IBM Engineering Lifecycle

Agent Skill for editing prose so it does not read like generic LLM output

an ai based agent writing with go and ts, inspired by Claude Code.

AI-Agent-first architecture spec-writing framework + Claude Code plugin harness (Layer 0 META FRAMEWORK). HTML-only specs, 2-layer (folio + consumer), self-hosting.

Convert Claude Code conversations into a searchable markdown knowledge base using the Claude Agent SDK without vector databases or complex RAG setups.

与吉伊卡哇世界的角色们对话！包含乌萨奇、小可爱、小八、小桃、铠甲人三兄弟、狮萨、海獭、栗子馒头共 10 位角色设定，单一入口按需加载，支持实时切换。适配任何 Markdown Skill/Prompt 格式的 AI Agent 系统。

Agent Python qui documente automatiquement du code : génération de docstrings via LLM (Claude / Ollama), injection AST, README et CHANGELOG automatiques. TP6 série Developer Tooling IA

Multi-agent research orchestrator built with the Claude Agent SDK. A coordinator spawns parallel web-search and document-analysis subagents, merges their structured findings, then drives a synthesis agent to produce a fully-cited Markdown report. Demonstrates parallel fan-out, structured output contracts, and attribution verification.

A CrewAI-powered news aggregator that uses a team of AI agents (Researcher, YouTube Scout, Summarizer) to fetch, analyze, and summarize the latest AI news. Runs with local LLMs via Ollama. Outputs a markdown report and TTS audio via a FastAPI web server.

可直接安装给codex的skill，把学术 PDF 自动整理成中文深读报告：摘要、结构、定理证明、图表分析、参考文献，并生成 Markdown/TeX/PDF，更好地辅助科研工作者对文章进行理解

Remote Model Context Protocol (MCP) server for Squad AI. Manage your product strategy with your favourite AI agent.

Thisis an intelligent agent system powered by LLM tool calling with a revolutionary "Skill-First" architecture . It transforms business rules into Markdown documents, enables on-demand Skill loading and multi-Skill composition, and executes atomic commands within a secure sandboxed workspace.

Files-first operating system for internet businesses: research, decisions, bets, pushes, and outputs in markdown/git, with an mb CLI and agent workflows that hide the plumbing.