159 boosters for "local" — open source, verified from GitHub, ready to install
A web search MCP server powered by Brave Search API, enabling Claude and other AI tools to search the web, find local businesses, and retrieve images in real-time. Useful for developers building AI applications that need current information beyond training data.
"name": "claude-blog", "description": "AI-powered blog skill suite with 30 sub-skills and 5 agents. FLOW framework integration (Find/Optimize/Win, 30 evidence-led prompts), semantic topic-cluster planning + execution, multilingual publishing (translate/localize/locale-audit), Google API integration
"description": "Durable, fully-local project memory for Claude Code. Logs every session to .recall/history.md and condenses it into .recall/context.md with a local Python summarizer (TF-IDF + TextRank) — no API key, no external model, nothing leaves your machine. New sessions resume from the saved c
Use this skill when the user wants a repeatable, auditable implementation workflow for a non-trivial coding task, especially a feature, refactor, migration, or bug fix that should leave repo-local proof in . All task artifacts created by this workflow must stay inside the repository. When the exampl
<!-- This file is derived from CLAUDE.md. Keep them in sync. --> A lightweight, file-backed vector database for Node.js and browsers with Pinecone-compatible filtering and hybrid BM25 search.
Quoroom is an experimental open-source framework for building local AI agent systems with a Queen-Workers-Quorum architecture, designed for researchers and developers exploring multi-agent AI patterns with Claude and other LLMs.
Build robust native Android applications using Kotlin with modern architecture patterns, Jetpack libraries, and Compose for declarative UI.
Use this skill only for explicit or requests. is the only supported entrypoint.
You're a patient mentor helping a complete beginner turn a fuzzy app idea into something concrete they can actually build, and stay calm while they build it. You're not an interrogator. You're the friend who's done this before, sitting next to them on their first flight. Your job is to help them fin
This skill helps optimize your local development workflow with Cursor IDE. It provides guidance on daily development patterns, keyboard-driven workflows, and efficient AI-assisted coding practices to maximize productivity. 1. Open project and let indexing complete 2. Review recent changes with git d
Regenerate in the project root: a complete point-in-time snapshot of everything open — PRs (grouped by whose move it is), local branches without PRs, issues (grouped by theme, with reply/duplicate/
AgentRecall is a persistent memory system. Default surface: 5 tools (two verbs + three essentials). Full surface: 18 tools via . This guide describes how and when to use them. AgentRecall requires the MCP server to be running. If tool calls fail with "unknown tool", the human needs to install it fir
This skill provides comprehensive guidance for all aspects of Playwright test development, from writing new tests to debugging and maintaining existing test suites. Consult these references based on what you're doing:
Use this skill as a global-installable, project-level rebuttal workspace assistant. The installed skill provides reusable procedures and assets; each actual paper/rebuttal workspace gets its own state folder for memory, snapshots, template state, and logs. Start by understanding the workspace, then
Automate reproducible marketing screenshot capture for a SwiftUI iOS app across multiple locales, with two parallel output streams: 1. Full-screen captures — every marketing-relevant screen, with deterministic seeded data, real status bar / safe-area chrome 2. Element captures — isolated renders of
Smart Coding MCP enhances developer productivity with AI-powered semantic code search, intelligent indexing, and hybrid search capabilities integrated into Cursor, Claude Desktop, and VS Code. It's designed for developers working with local LLMs who need faster, smarter code discovery and navigation.
Use this skill for local, editable thesis defense decks that must follow a supplied university/lab/company PowerPoint template. 2. Extract thesis context: 3. Analyze the template:
"name": "readme-skill", "description": "Generate an AI-Native developer profile or monthly/time-window report from local Claude Code + Codex CLI + Kiro (AWS) + Trae (ByteDance) + Gemini Antigravity (Google) + Cursor composer metadata, GitHub contributions, and git log. Local-only, read-only, anonymi
"description": "Run any model with an Anthropic- or OpenAI-compatible API (e.g. DeepSeek, GLM, Kimi, Qwen, MiniMax) — even your Codex subscription — as real Claude Code workflows, agent-team teammates, or one-shot subagents, driven exactly like native ones. Your main session's own auth is untouched
Generate XML files and export to PNG/SVG/PDF/JPG locally using the native draw.io desktop app CLI. PNG, SVG, and PDF exports support () — the exported file contains the full diagram XML, so opening it in draw.io recovers the editable diagram. Use double extensions () to signal embedded XML. The dr
"description": "Nemp Memory — Cognitive memory layer for AI agents across every platform. Local-first, zero cloud, works with Claude, Codex, Cursor, and Windsurf.", "name": "Sukin Shetty", "email": "contact@nemp.dev",
DyNote 是面向 Codex 这类 Agent 的抖音学习工具，不是一次性摘要器。核心原则是“数据资产先行，学习笔记后置”：先把字幕/转写、评论、元数据和 AI 快读沉淀为可复用资产，再按用户需求生成可追溯的学习笔记、总结和写作材料。默认目标不是字幕工程文件，而是先落一份原始数据包：、、、、、、，并用 归档可复用资产。默认把独立字幕轨或本地自动语音识别转写当作事实主干；当转写密度低、任务需要画面理解或用户只要快速筛选时，再用已登录抖音网页版的“问AI / 识别画面”补充，豆包只作为抖音 AI 不可用时的备用快读或待核验假设。 联网或登录态操作必须先使用 。不要读取、复制或打印 Cooki
This skill should be used when users want to train or fine-tune language models using TRL (Transformer Reinforcement Learning) on Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment. Includes guidance on the TRL Jobs package, UV scripts with PEP 723 format, dataset preparation and validation, hardware selection, cost estimation, Trackio monitoring, Hub authentication, and model persistence. Should be invoked for tasks involving cloud GPU training, GGUF conversion, or when users mention training on Hugging Face Jobs without local GPU setup.
A local SQLite-based command-line task board for AI agents and developers to manage multi-step coding tasks, track progress, and coordinate work across sessions without external dependencies. Ideal for breaking down complex implementations into trackable subtasks with comments and checklists.
