qingni

qingni / AgentCrew

Public

A universal orchestration workbench for macOS. Seamlessly mix AI models (Codex, Claude) with traditional CLI tools (git, npm) via static DAG pipelines or dynamic self-healing agents, creating a fully automated closed-loop for development and ops.

ai-agents automation cli dag developer-too
16
1
100% credibility
Found Mar 19, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Swift
AI Summary

AgentCrew is a macOS app for creating visual workflows that combine AI assistants and regular computer commands to handle development, testing, and automation tasks locally.

How It Works

1
🖥️ Discover AgentCrew

You hear about this handy Mac app that teams up smart helpers to automate your coding chores without hassle.

2
🚀 Launch on your Mac

Open the app and it greets you warmly, checking if your helpers are ready to go.

3
📁 Pick your project spot

Choose the folder with your work so everything happens right where your files are.

4
✨ Describe your goal

Type what you want like 'add user login and tests' – the app smartly builds a step-by-step plan.

5
Tweak the plan
🤖
Let AI lead

Trust the auto-plan and jump to running.

✏️
Make your own

Add or change steps to fit just right.

6
▶️ Hit play and watch

Choose simple run or smart retry mode, then see steps flow with live updates and alerts.

✅ Task done beautifully

Your code is updated, tested, and ready – celebrate the smooth win with full logs to review.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 16 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is AgentCrew?

AgentCrew is a macOS-native SwiftUI app for universal agentic orchestration, letting you chain AI tools like Codex, Claude, and Cursor with any CLI commands—git, npm, docker, ffmpeg—in visual DAG pipelines or self-healing agent loops. It solves the mess of manual scripting by creating automated closed-loop workflows for dev tasks like code gen, review, testing, and ops like batch media processing or self-healing checks. Fire up a natural language prompt, and it auto-plans the steps, runs them with concurrency waves, and handles retries or human approval.

Why is it gaining traction?

It stands out by mixing AI models and traditional CLI tools in one workbench, with dual modes—fast static pipelines for predictable tasks or dynamic agents that diagnose failures and replan—plus real-time monitoring, mode insights dashboards, and one-click CLI profile switching. Developers dig the AI auto-planner that turns "add JWT auth and tests" into executable flows, wave-based concurrency for speed, and exportable logs for tuning. No more juggling terminals and half-baked scripts.

Who should use this?

MacOS devs building local CI/CD pipelines, like running lint/test/build in waves before AI-generated changelogs. AI tinkerers chaining Claude for analysis, Cursor for reviews, Codex for fixes in long dev loops. Ops folks automating ffmpeg batches or service self-healing with CLI + agents.

Verdict

Worth a spin for macOS users with those AI CLIs—solid docs, demo pipelines, and Apache license make it easy to eval. At 16 stars and 1.0% credibility, it's early and unproven; test on non-critical workflows until more runs validate reliability.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.