Cloud AI — General Coding + Data + Data Science + AI Engineering

A Vibe Coding assistant for VS Code that goes far beyond simple chat. Cloud AI turns your editor into a specialist across four full areas:

• General coding — explain, refactor, generate, fix code in any language
• Data analytics — inventory datasets, frame business questions, drill into root causes, generate decks
• Data Science — clean, EDA, build/train/tune ML models, explain them with SHAP/LIME
• AI Engineering — production RAG, prompt engineering, LoRA fine-tuning, LLM evaluation, agents

All powered by large open-weight models (Qwen, Gemma, custom GGUFs) running on Google Colab Pro GPUs, streamed to your VS Code over a Cloudflared tunnel. No local GPU required.

Architecture in one diagram

   ┌─────────────────────┐               ┌──────────────────────────┐
   │ VS Code (you)       │               │ Google Colab Pro (L4/A100)│
   │  ┌───────────────┐  │   HTTPS/SSE   │  ┌────────────────────┐  │
   │  │ Cloud AI      │◄─┼───tunnel──────┼─▶│ FastAPI router     │  │
   │  │ sidebar chat  │  │               │  │  + Ollama runtime  │  │
   │  └───────────────┘  │               │  └────────────────────┘  │
   │  - 14 slash skills  │               │    Qwen 27B · Gemma 26B  │
   │  - Tool calling     │               │    (or your own GGUF)    │
   │  - Voice + vision   │               │                          │
   └─────────────────────┘               └──────────────────────────┘

The big models live where there's VRAM. Your editor stays light. Tools execute only on your machine — the backend just emits intents.

What you can do — the four areas

🧑‍💻 General coding (works everywhere)

• Open Chat: dedicated sidebar chat with a Cloud AI icon in the Activity Bar (same UX as Copilot Chat / Claude Code). Click and use, no setup ritual.
• Inline commands:
  • Cloud: Explain Selection (Ctrl+Alt+E)
  • Cloud: Refactor Selection (Ctrl+Alt+R)
  • Cloud: Generate at Cursor
• Tool calling: the model can read/write files, run bash, grep, glob, fetch the web, etc. Every tool call shows a confirmation modal with Accept / Always / Deny. Dangerous bash patterns (rm -rf /, dd, fork bombs) block the "Always" option as a safety net.
• Voice input and image attachments for vision-capable models.
• Persistent chat history across sessions.

📊 Data analytics — /explore, /question, /chart, /drill, /report

Adapted from ai-analyst-lab/ai-analyst.

🔬 Data Science (ML) — /engineer, /eda, /model, /explain

Adapted from HungHsunHan/claude-code-data-science-team.

🤖 AI Engineering — /rag, /prompt, /finetune, /eval, /agent

Original to Cloud AI — built from current best practices.

Type /skills or /help in the chat to list all 14 with descriptions.

Quick start

1. Install — download the latest cloud-ai-*.vsix from this repository's Releases (or build it yourself: npm install && npm run package), then:

   code --install-extension cloud-ai-0.5.0.vsix
   

2. Spin up the backend — open backend/cloud_colab.ipynb in Colab (one-click), pick a GPU runtime and run all cells. The last cell prints:

   CLOUD_URL=https://xxxxx.trycloudflare.com
   CLOUD_TOKEN=...
   

3. Configure VS Code — open Settings → search "Cloud AI":
   • cloud.tunnelUrl → the Cloudflared URL
   • cloud.token → the token
4. Click the Cloud AI icon in the Activity Bar (the brain + chip in the left rail) — the chat opens in the sidebar.
5. Type /skills to see everything you can do.

Running the backend somewhere other than Colab

The backend is a small FastAPI server over Ollama — see backend/README.md. Any machine with Ollama works:

pip install -r backend/requirements.txt
CLOUD_TOKEN=change-me python backend/server.py      # then set cloud.tunnelUrl=http://localhost:8000

It also documents the wire protocol (/health, /vibe/ask, /vibe/chat over SSE) if you want to point the extension at your own inference stack.

Settings

Keybindings

Privacy & security

• Tools require explicit per-session approval. "Always" decisions live in RAM and reset when VS Code closes.
• Token is machine-scoped — does not sync across devices via Settings Sync.
• All traffic flows through your own Cloudflared tunnel and your own Colab session. No third-party server in the middle.
• File system access and shell execution happen only on your machine — the backend just forwards the model's intent.

Why this stack?

• VRAM economics. Qwen 27B and Gemma 26B don't fit on an 8GB local GPU. They fit on Colab Pro's L4 / A100.
• Zero per-token cost. Colab Pro is a flat monthly fee — vs. metered API billing.
• Privacy. Prompts never leave a Colab kernel you control.
• Editor stays local. All file I/O and shell execution happen on your machine.

Limitations

• Needs a Colab Pro account (or another GPU host) to run the backend. The extension does not include hosted inference.
• The Cloudflared tunnel URL changes when Colab restarts — you must update the setting (the extension shows a clear error when it's offline).
• AI Engineering skills generate code; they don't run training/RAG/eval pipelines for you. They tell you what to write and why.

Development

npm install            # esbuild + TypeScript + vsce, no runtime dependencies
npm run typecheck      # tsc --noEmit
npm run build          # bundles src/ -> dist/extension.js
npm run package        # produces cloud-ai-<version>.vsix
npm run test:backend   # protocol smoke test against a mock Ollama (needs backend/requirements.txt)

Press F5 in VS Code to launch an Extension Development Host with the extension loaded.

Issues & contributing

Bug reports, feature requests, and PRs are welcome at github.com/cloud-fronts-group/cloud-ai.

Provenance

Cloud AI began as Mendes AI, written by Alex Mendes and published to the VS Code Marketplace as mendesalex89.mendes-ai (v0.1.0–0.3.2, May 2026). The source repository the listing links to, mendesalex89/AI_Assistant, no longer exists on GitHub, and none of the author's other public repositories contain it.

This repository restores the project:

• Extension source (src/, media/, skills/) is the original 0.3.2 TypeScript, recovered verbatim from the dist/extension.js.map source map that shipped inside the published VSIX. The only functional change is a spec-compliant SSE parser in src/client.ts (see the changelog).
• Backend (backend/) is a reconstruction: the original Colab notebook was lost with the repo, so server.py re-implements the protocol the client speaks, on Ollama, plus a new Colab notebook.
• Build tooling (esbuild.config.mjs, tsconfig.json, .vscodeignore, CI) was not in the VSIX and was written fresh.

Everything is MIT, as the original was.

Acknowledgments

Skills are adapted from these excellent open-source projects:

• Data analytics skills from ai-analyst-lab/ai-analyst (Apache 2.0)
• Data Science skills from HungHsunHan/claude-code-data-science-team

License

MIT — see LICENSE. Original work © 2026 Alex Mendes; source recovery, backend and tooling © 2026 Cloud Fronts Group.