Joseph Steinbrenner

I like to build things

[Joseph Steinbrenner @josstei_dev]

maestro-orchestrate — codex-skill by josstei https://t.co/8URSIKGWsx

🏅 Ranked #5 for Quality out of all Codex Skills!
🏅 Ranked #29 for Quality out of ALL for repos with 100+ stars!

Multi-agent orchestration platform for Gemini CLI, Claude Code, Codex, and Qwen Code — 39 specialists, parallel subagents, persistent sessions, and built-in code review, debugging, security, SEO, accessibility, and compliance tools

Give Maestro a try and build something awesome!
https://t.co/fbioEBrcWM

#Codex #GeminiCli #ClaudeCode #QwenCode #Coding #OpenAI #Anthropic

[Joseph Steinbrenner @josstei_dev]

@london_joh23108 Sounds good with me. I’ll keep an eye out for the PR

[Joseph Steinbrenner @josstei_dev]

17 new specialist agents.

Qwen Code runtime.

Server-side design approval gates.

maestro-orchestrate v1.6.3 is out — one prompt, a team of 39 specialists coordinating your CLI coding work.

https://t.co/kNdDU2uHA3 ↓

[Joseph Steinbrenner @josstei_dev]

maestro-orchestrate — codex-skill by josstei https://t.co/8URSIKGWsx

🏅 Ranked #5 for Quality out of all Codex Skills!
🏅 Ranked #29 for Quality out of ALL for repos with 100+ stars!

Multi-agent orchestration platform for Gemini CLI, Claude Code, Codex, and Qwen Code — 39 specialists, parallel subagents, persistent sessions, and built-in code review, debugging, security, SEO, accessibility, and compliance tools

Give Maestro a try and build something awesome!
https://t.co/fbioEBrcWM

#Codex #GeminiCli #ClaudeCode #QwenCode #Coding #OpenAI #Anthropic

[Joseph Steinbrenner @josstei_dev]

17 new specialist agents.

Qwen Code runtime.

Server-side design approval gates.

maestro-orchestrate v1.6.3 is out — one prompt, a team of 39 specialists coordinating your CLI coding work.

https://t.co/kNdDU2uHA3 ↓

[Joseph Steinbrenner @josstei_dev]

Full changelog + install Instructions: https://t.co/kNdDU2uHA3

#VibeCoding #ClaudeCode #GeminiCLI #Codex #BuildInPublic

[Joseph Steinbrenner @josstei_dev]

What's new:

→ Roster 22 → 39: ML/AI, SRE, platform, mobile, mainframe (COBOL, DB2, zOS, HLASM, IBM i)
→ Qwen Code runtime — 4th supported CLI
→ Design gates enforced before execution
→ Phase reconciliation when agents drift from plan

[Joseph Steinbrenner @josstei_dev]

Maestro just went multi-runtime. 🚀

Gemini CLI. Claude Code. OpenAI Codex.

Pick your weapon.

What is it? You describe what you want to build. Maestro spins up a full dev team — 22 AI specialists — and ships it. Architect designs it. Coder builds it. Tester tests it. Security engineer audits it. Code reviewer signs off. All running in parallel. All coordinated by a TechLead that won't let broken code through.

One command. Real software. No babysitting.

And now it works on whichever AI coding tool you actually use.

v1.6 is LIVE.

294 stars and climbing.

https://t.co/fbioEBqF7e

#VibeCoding #ClaudeCode #GeminiCLI #Codex #BuildInPublic

[Micha(el) Bladowski 🇩🇪 🇺🇦 @michabbb]

#Maestro: Multi-Agent Orchestration for #GeminiCLI & #ClaudeCode #AIAgents #opensource

🤖 22 specialized subagents across 8 domains — Engineering, Product, Design, Content, SEO, Compliance, I18n & Analytics — directed by a TechLead orchestrator

��� Two workflow modes:
Express — simple tasks, 1-2 questions, single-agent delegation, code review, archival
Standard — full 4-phase: Design Dialogue, Planning, Execution, Completion with quality gate

🔀 Native parallel execution: independent phases dispatched as concurrent subagent batches with non-overlapping file ownership — no waiting, no blocking

🛡️ Least-privilege 4-tier security model:
Read-Only → Architect, Code Reviewer
Read+Shell → Debugger, Security Engineer
Read+Write → Technical Writer, Product Manager
Full Access → Coder, Tester, DevOps

📋 Standalone commands without full orchestration:
/review /debug /security-audit
/perf-check /seo-audit /a11y-audit /compliance-check

🔄 Persistent session state in YAML+Markdown:
📌 Phase tracking & file manifests
📌 Token usage per agent
📌 Error logs & retry counts
📌 Reliable resume after interruptions

🏗️ Guided design dialogue with depth-selectable reasoning (Quick/Standard/Deep), built-in codebase investigation, and 2-3 architectural approaches with trade-offs before any code is written

✅ Final quality gate blocks completion on unresolved Critical/Major findings — remediates and re-validates until clean

🔧https://t.co/tIAUaYae8A

[Joseph Steinbrenner @josstei_dev]

Maestro just went multi-runtime. 🚀

Gemini CLI. Claude Code. OpenAI Codex.

Pick your weapon.

What is it? You describe what you want to build. Maestro spins up a full dev team — 22 AI specialists — and ships it. Architect designs it. Coder builds it. Tester tests it. Security engineer audits it. Code reviewer signs off. All running in parallel. All coordinated by a TechLead that won't let broken code through.

One command. Real software. No babysitting.

And now it works on whichever AI coding tool you actually use.

v1.6 is LIVE.

294 stars and climbing.

https://t.co/fbioEBqF7e

#VibeCoding #ClaudeCode #GeminiCLI #Codex #BuildInPublic

[Joseph Steinbrenner @josstei_dev]

🚀 **Maestro v1.5.0 is already rewriting the rules.**

Maestro is an orchestration layer that finally makes Gemini CLI’s native subagents and Claude Code’s plugin system *play together perfectly*.

You describe the outcome → Maestro classifies complexity, runs the design dialogue, builds the dependency graph, then unleashes specialized subagents in parallel — coder, architect, tester, security engineer, debugger — all under hard quality gates and least-privilege enforcement.

Read-only agents literally *cannot* write files.
Shell agents can’t touch your code.
Everything runs on *one codebase*, two runtimes, zero drama.

This is what production-grade agentic development actually looks like.

If you haven’t installed it yet:

** Gemini CLI

gemini extensions install https://t.co/fbioEBqF7e

** Claude Code
claude plugin marketplace add josstei/maestro-orchestrate

claude plugin install maestro@maestro-orchestrator --scope user

Drop what you’re building in the replies!

The agentic future isn’t coming — it’s already here, and it’s open source.

Let’s go build. 🔥

#AgenticAI #GeminiCLI #ClaudeCode #AICoding #DevTools

[Joseph Steinbrenner @josstei_dev]

@ihtesham2005 I wonder how maestro-orchestrate would work in your tool, nested orchestration layers seem powerful.

https://t.co/QtQ5sQPZ1e

[Joseph Steinbrenner @josstei_dev]

🚀 **Maestro v1.5.0 is already rewriting the rules.**

Maestro is an orchestration layer that finally makes Gemini CLI’s native subagents and Claude Code’s plugin system *play together perfectly*.

You describe the outcome → Maestro classifies complexity, runs the design dialogue, builds the dependency graph, then unleashes specialized subagents in parallel — coder, architect, tester, security engineer, debugger — all under hard quality gates and least-privilege enforcement.

Read-only agents literally *cannot* write files.
Shell agents can’t touch your code.
Everything runs on *one codebase*, two runtimes, zero drama.

This is what production-grade agentic development actually looks like.

If you haven’t installed it yet:

** Gemini CLI

gemini extensions install https://t.co/fbioEBqF7e

** Claude Code
claude plugin marketplace add josstei/maestro-orchestrate

claude plugin install maestro@maestro-orchestrator --scope user

Drop what you’re building in the replies!

The agentic future isn’t coming — it’s already here, and it’s open source.

Let’s go build. 🔥

#AgenticAI #GeminiCLI #ClaudeCode #AICoding #DevTools

[Joseph Steinbrenner @josstei_dev]

Gemini CLI shipped native subagent support. Claude Code has had it.

Maestro v1.5.0 orchestrates both — 22 specialized subagents across 8 domains, dispatched in parallel with least-privilege tooling.

Describe what you want to build. Maestro classifies complexity, runs a design dialogue, generates an implementation plan with dependency graphs, then delegates to the right subagents — coder, architect, tester, security engineer, debugger — automatically.

Hard-gated quality checks. Agent capability enforcement. Read-only agents can't write files, shell-only agents can't touch code. Parallel execution for independent phases.

One codebase. Two runtimes. Same workflow.

Gemini CLI

gemini extensions install https://t.co/fbioEBqF7e

Claude Code

claude plugin marketplace add josstei/maestro-orchestrate
claude plugin install maestro@maestro-orchestrator --scope user

#GeminiCLI #ClaudeCode #AI #OpenSource #AIAgents #DevTools #AICoding #LLM #BuildInPublic #SoftwareEngineering #Automation #AgenticAI #CodingTools #Developer

[Ajitesh Shukla @vitalflux]

Multi-agent orchestration just stopped being a toy. Maestro running on both Claude Code and Gemini CLI means you can actually coordinate 22+ specialized agents without building your own control plane.

Most teams still treat agents like chatbots with tools. Maestro forces you to think like an actual system architect—state management, parallel execution, quality gates. That's the mindset shift that matters.

The real play: orchestration platforms are eating custom prompt engineering. If you're still chaining API calls with retry logic in application code, you're doing 2023 work.

#AgenticAI #LLM

https://t.co/kXiKt2vCmW

[Joseph Steinbrenner @josstei_dev]

If you want to fully leverage all that sub-agents have to offer give Maestro a try!

Sub-agent orchestration platform for Gemini CLI and Claude Code — 22 specialists, parallel subagents, persistent sessions, and built-in code review, debugging, security, SEO, accessibility, and compliance tools

https://t.co/fbioEBqF7e

[Joseph Steinbrenner @josstei_dev]

Gemini CLI shipped native subagent support. Claude Code has had it.

Maestro v1.5.0 orchestrates both — 22 specialized subagents across 8 domains, dispatched in parallel with least-privilege tooling.

Describe what you want to build. Maestro classifies complexity, runs a design dialogue, generates an implementation plan with dependency graphs, then delegates to the right subagents — coder, architect, tester, security engineer, debugger — automatically.

Hard-gated quality checks. Agent capability enforcement. Read-only agents can't write files, shell-only agents can't touch code. Parallel execution for independent phases.

One codebase. Two runtimes. Same workflow.

Gemini CLI

gemini extensions install https://t.co/fbioEBqF7e

Claude Code

claude plugin marketplace add josstei/maestro-orchestrate
claude plugin install maestro@maestro-orchestrator --scope user

#GeminiCLI #ClaudeCode #AI #OpenSource #AIAgents #DevTools #AICoding #LLM #BuildInPublic #SoftwareEngineering #Automation #AgenticAI #CodingTools #Developer

[Joseph Steinbrenner @josstei_dev]

Gemini CLI is one AI doing everything.

Maestro splits it into 22 specialists — architect, coder, debugger, security engineer, SEO, a11y, compliance — each with its own context, tools, and least-privilege access.

A TechLead orchestrator classifies your task, picks the right workflow, delegates phases, and runs them in parallel using Gemini's native subagent scheduler.

Simple task? Express path — 1-2 questions, a brief, one agent, done.
Complex task? Full 4-phase workflow — design dialogue, implementation plan, parallel execution, quality gate.

It's a Gemini CLI extension.

install:
gemini extensions install https://t.co/s397TYL8jV

https://t.co/lQxgvQmCNU

#GeminiCLI #AIAgents #OpenSource