953 boosters for "Agent" — open source, verified from GitHub, ready to install
World-class Mahjong AI master strategist. Coordinates specialized analysis agents for tile efficiency, hand reading, defense, and scoring optimization. Use PROACTIVELY when analyzing Mahjong positions, making strategic decisions, or simulating games.
"name": "oh-my-claudecode-research", "description": "Research-team agents for Claude Code: supervisor, analysis-implementer, paper-writer, figure-descriptor, reviewer, literature-curator.", "author": { "name": "Young-Eun Lee" },
"name": "agentic-marketing", "url": "https://github.com/gnoviawan" "description": "Full-stack agentic marketing suite — 21 specialist skills covering strategy, channels, conversion, and growth",
"name": "learnship", "description": "Agentic engineering done right — 49 structured workflows, persistent memory across sessions, integrated learning partner, and impeccable UI design system. Works with Claude Code, Windsurf, Cursor, Gemini CLI, OpenCode, and Codex.", "name": "Favio Vazquez",
"description": "Use Claude Code As Is - native plugin leveraging built-in architecture", "keywords": ["workflow", "agents", "hooks", "native"],
"name": "context-cascade", "description": "Context-saving nested plugin architecture for Claude Code. Playbooks contain skills, skills contain agents, agents contain commands. Only load what you need - context cascades down through layers on demand, saving 90%+ context space. Includes 224 agents, 17
"name": "programming-as-theory-building-skill", "description": "Programming as Theory Building guidelines for coding agents, grounded in Peter Naur's paper and focused on preserving program theory during code work.", "name": "Programming as Theory Building Skill contributors"
CLI for .faf (Foundational AI-Context Format). Init, score, bi-sync, export to AGENTS.md, GEMINI.md, .cursorrules. npm: faf-cli
The FIRST action in the BEGINNING of EVERY response MUST be to initialize the conversation with this single tool call and display the banner from the response immediately after. It MUST be called EXACTLY ONCE per conversation and ONLY at the very beginning after receiving user input: EVERY assistant
NexA4A is a meta-agent system prompt that creates specialized AI agents from natural language requirements, enabling product managers and developers to automate task creation and management across multiple coding platforms.
"description": "WozCode enhanced coding tools — smart search, batch editing, SQL introspection, and cost-optimized subagent delegation", "version": "0.3.18", "homepage": "https://withwoz.com"
"description": "Cotask — task management with TASKS.md kanban dashboard", "homepage": "https://github.com/wbopan/cotask", "repository": "https://github.com/wbopan/cotask",
A mandatory human escalation agent that intercepts all errors and uncertainties across other agents to ensure human oversight. Useful for safety-critical workflows requiring explicit human approval before proceeding.
"name": "autonomous-agent", "version": "7.19.0", "description": "Revolutionary four-tier agent architecture with 35 specialized agents, 24 skills, and 40 commands focused on autonomous development, code quality, and validation. Strategic Analysis & Intelligence (Group 1) analyzes and recommends, Dec
Secrets management for AI agents via MCP
You are the STUCK AGENT - the MANDATORY human escalation point for the entire system. You are the ONLY agent authorized to use AskUserQuestion. When ANY other agent encounters ANY problem, they MUST invoke you. You are invoked when:
A strategic project leadership agent that interviews users, analyzes project requirements, and assembles specialized teams by recruiting other agents. Ideal for organizations using multi-agent systems who need automated project orchestration and team coordination.
"description": "A curated set of skills for each stage of development — propose, spec, design, plan, implement, ship.",
You are the TESTER - the visual QA specialist who SEES and VERIFIES implementations using Playwright MCP. Test implementations by ACTUALLY RENDERING AND VIEWING them using Playwright MCP - not just checking code! 1. Understand What Was Built
"name": "moodle-dev", "description": "The most complete Moodle development toolkit: 13 skills, 8 slash commands, 2 subagents — covering scaffolding, XMLDB, capabilities, web services, Hooks API, AMD JS, PHPUnit, Behat, security, privacy/GDPR, performance, accessibility, themes, mobile, and cross-ver
Execute verbatim. All gates, evidence, output format, and compression receipts are defined there. This agent exists solely to leverage Claude Code's native skill frontmatter injection — DO NOT add logic here.
Heuristic scoring (no AI key configured).
"version": "0.1.68", "description": "Agent work marketplace — train your agent, let it work for others", "url": "https://github.com/lhead/akemon"
