mrgoonie

Human-friendly & AI agent-friendly CLI for Human MCP — vision, speech, generation & reasoning in your terminal

26
3
100% credibility
Found Apr 15, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

human-cli is a terminal tool that provides AI-driven capabilities for image analysis, media generation, text-to-speech, and reasoning directly from the command line.

How It Works

1
📰 Discover human-cli

You hear about a handy terminal tool that lets your computer see images, create art and videos, speak words aloud, and think through problems like a smart friend.

2
📥 Install easily

With one quick command, you add this superpower kit to your computer's terminal toolbox.

3
🔗 Connect the AI brain

You share a private pass with a clever AI service, like Gemini, so the tool can borrow its smarts for seeing, creating, and talking.

4
🩺 Check if ready

Run a simple health check to confirm all the eyes, hands, mouth, and brain features light up perfectly.

5
Use the magic powers

Point it at a photo to spot issues, dream up images or videos from words, hear text read aloud in fun voices, or brainstorm solutions step by step—it's thrilling!

6
Choose your adventure
🙋
Solo fun

Jump right in with simple commands to create, analyze, and explore on your own.

🤖
With AI friends

Hook it to smart agents like Claude so they can see, build, and speak through your terminal.

🎉 Terminal transformed

Your everyday command line now feels like a creative studio and thinking partner, making work playful and powerful.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 26 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 human-cli?

human-cli is a TypeScript CLI that packs vision, speech, image/video/audio generation, and reasoning into your terminal, powered by Gemini and optional providers like Minimax and ElevenLabs. It solves the hassle of juggling separate AI APIs for quick prototyping—run `human eyes analyze screenshot.png` for UI bugs, `human hands gen-image "red fox in snow"` for assets, or `human brain think "rate limiter design"` for step-by-step logic. Human-friendly with spinners and colors, it's also agent-friendly via JSON envelopes and MCP server mode for Claude Desktop.

Why is it gaining traction?

No subprocesses mean 30x faster starts than older MCP wrappers, with native tool registry for direct calls like `human call eyes_analyze --args '{"source":"img.png"}'`. Agents love stable exit codes, stdin piping, and `human tools --json` discovery; humans get sensible outputs in `./outputs/`. Optional deps keep installs lean, and Claude Code plugins auto-teach usage.

Who should use this?

Terminal hackers prototyping AI vision for docs/images, AI agent devs building shell skills for Claude Cowork or Antigravity, or Claude Desktop users needing local gen/reasoning without browser tabs. Ideal for solo devs scripting human clipart generation or speech narration in workflows.

Verdict

Grab it for fast terminal AI experiments if you're in the Claude ecosystem—v2.0 rewrite shines on speed—but 14 stars and 1.0% credibility signal early maturity; test thoroughly before production. (187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.