gemini-kit and maestro-gemini

The first is a broader CLI agent framework with 19 general-purpose agents and 44 commands, while the second is a specialized multi-agent orchestration platform built on top of Gemini CLI's foundation, making them complements that can work together with maestro-gemini providing advanced coordination capabilities for gemini-kit's agents.

gemini-kit
51
Established
maestro-gemini
40
Emerging
Maintenance 13/25
Adoption 10/25
Maturity 13/25
Community 15/25
Maintenance 13/25
Adoption 9/25
Maturity 11/25
Community 7/25
Stars: 273
Forks: 28
Downloads:
Commits (30d): 0
Language: Shell
License: MIT
Stars: 103
Forks: 4
Downloads:
Commits (30d): 0
Language: JavaScript
License: MIT
No Package No Dependents
No Package No Dependents

About gemini-kit

nth5693/gemini-kit

🚀 19 AI Agents + 44 Commands for Gemini CLI - Code 10x faster with auto planning, testing, review & security

Built on Google's Gemini API, Gemini-Kit implements a multi-agent orchestration architecture where specialized agents (Security Auditor, Frontend Specialist, Backend Specialist, etc.) coordinate through a central Orchestrator agent to handle complex development tasks. The system features a compound workflow pattern (`/explore → /plan → /work → /review → /compound`) that builds a persistent knowledge base from each iteration, enabling the AI to learn from corrections and reuse solutions. It integrates with version control via Git hooks for automatic checkpoints, includes 30+ secret detection patterns for real-time security blocking, and supports Discord/Telegram notifications for external alerting. --- **Word count: 80 | Sentence count: 3**

About maestro-gemini

josstei/maestro-gemini

Turn Gemini CLI into a multi-agent platform — 12 specialized subagents, parallel dispatch, 4-phase orchestration, and standalone dev tools for code review, debugging, security, and performance

Extends Gemini CLI with 22 domain-specialist subagents orchestrated through complexity-adaptive workflows (Express for simple tasks, Standard 4-phase for medium/complex), using native parallel execution, persistent YAML+Markdown session state, and MCP-first operations. Includes standalone audit commands (code review, security, performance, SEO, accessibility, compliance) with least-privilege tool access enforced through agent frontmatter and 4-tier permission tiers.

Scores updated daily from GitHub, PyPI, and npm data. How scores work