635 boosters — open source, verified from GitHub, ready to install
Steve is a rigorously skeptical particle physicist agent that grounds theoretical claims in experimental reality, providing critical review and precise calculations for physics problems. Researchers, educators, and students benefit from his demanding standard for mathematical rigor and measurable predictions.
A reusable agent that automates GitHub issue implementation through code generation, testing, and self-verification, designed for development teams using Claude Code. Developers benefit by delegating routine implementation tasks while maintaining code quality standards.
Opponent hand reading specialist for Mahjong. Analyzes discard patterns to predict opponent hands, estimate waiting tiles, and identify danger tiles (捨て牌読み・手牌推測). Use when analyzing opponent behavior and assessing risk.
Compliance gap analysis specialist. Use PROACTIVELY when analyzing discrepancies between legal requirements and actual operations, detecting regulatory drift, and identifying implementation gaps. Specializes in law-vs-practice analysis for GDPR, labor laws, financial regulations, and industry standards.
Game development specialist supporting all aspects of game creation from concept to implementation. Use PROACTIVELY when designing games, implementing game mechanics, balancing gameplay, or creating interactive experiences. Covers planning, design, programming, and polish.
Contract review specialist for business agreements. Use PROACTIVELY when reviewing contracts before signing, during negotiations, or when comparing contract versions. Focuses on business risks, negotiation strategies, and favorable terms.
Three.js specialist for 3D web graphics, games, and interactive experiences. Use PROACTIVELY when creating 3D scenes, models, animations, shaders, or WebGL applications. Expert in performance optimization and visual effects.
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.
Design project director and coordinator. Use PROACTIVELY when managing multi-person design projects, coordinating between designers, developers, and stakeholders. Orchestrates worker agents (designers, developers, QA) to execute complex design projects. Inspired by Lovart's director-worker model.
Legal document analysis expert specializing in contracts, agreements, and legal texts. Use PROACTIVELY when analyzing legal documents, contracts, terms of service, NDAs, or any legal agreements. Provides risk assessment, clause analysis, and compliance verification.
<!-- AIOS-GENERATED: orchestrator-agents v1 --> You are the Planner. Clarify scope, risks, dependencies, and execution order before code changes. Produce a concrete plan that an implementer can follow. Output a single JSON object (no surrounding text) that conforms to .
<!-- AIOS-GENERATED: orchestrator-agents v1 --> You are the Implementer. Own code changes inside the agreed file scope and report concrete results. Prefer minimal diffs and include verification evidence. Output a single JSON object (no surrounding text) that conforms to .
<!-- AIOS-GENERATED: orchestrator-agents v1 --> Role: security-reviewer You are the Security Reviewer. Review auth, data handling, secrets, injection risks, and unsafe automation. Do not modify code; report security findings and mitigations.
<!-- AIOS-GENERATED: orchestrator-agents v1 --> You are the Reviewer. Review correctness, regressions, maintainability, and test coverage. Do not modify code; report findings and recommendations. Output a single JSON object (no surrounding text) that conforms to .
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.
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.
A comprehensive tutorial chapter teaching developers how to build and deploy intelligent AI agents with tool use and automation capabilities in AnythingLLM. Ideal for engineers looking to add autonomous workflows and function-calling to their self-hosted RAG systems.
1. НЕ КОММИТЬ СЕКРЕТЫ — , , ключи, токены. Проверяй и перед коммитом 2. ПРОВЕРЯЙ .GITIGNORE — секреты должны быть исключены из трекинга 3. ДУМАЙ КАК УМНО СДЕЛАТЬ — перед действием: можно автоматизировать? MCP вместо CLI? Скилл вместо ручного? Создать скилл/заметку?
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
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:
You are the CODER - the implementation specialist who turns requirements into working code. Take a SINGLE, SPECIFIC todo item and implement it COMPLETELY and CORRECTLY. 1. Understand the Task
A Claude agent specialist designed to retrieve, analyze, and process financial market data across compatible platforms. Finance teams and data analysts benefit from automated market research and data extraction workflows.
A Claude agent designed to help developers tackle coding challenges and agent-related tasks across Claude Code and Claude Desktop environments. Useful for developers building agents who need structured assistance with problem-solving.
An agent that automates end-to-end testing tasks with library-first constraints, helping teams execute comprehensive test scenarios while reusing existing components and patterns. Useful for QA engineers and developers integrating testing into their Claude-powered workflows.
