FrootAI for VS Code
From the Roots to the Fruits
20 Solution Plays 23 MCP Tools 16 Knowledge Modules 200+ AI Terms
What is FrootAI?
FrootAI is an AI architecture toolkit that gives you everything to build production AI directly inside VS Code. No web browser needed.
| What |
How |
| Browse 20 Solution Plays |
Pre-tuned AI solutions (RAG, agents, landing zones, voice AI) click to scaffold |
| Explore 23 MCP Tools |
Click any tool to see documentation, input/output, and examples |
| Search 200+ AI Terms |
Instant glossary lookup from the sidebar |
| Scaffold in One Click |
Init DevKit (19 files), TuneKit (8 configs), SpecKit (WAF alignment) |
| Estimate Azure Costs |
Pick a play + scale = monthly cost breakdown |
| Run Evaluations |
Auto-runs eval.py and shows quality dashboard |
Works standalone no need to clone the FrootAI repo. Install the extension and go.
Install
Option 1 VS Code Marketplace (recommended):
Ctrl+Shift+X Search "FrootAI" Install
Option 2 Terminal:
code --install-extension frootai.frootai-vscode
The extension adds a FrootAI icon to your activity bar with two panels:
FAI Solution Plays
Browse all 20 deployable AI solutions, organized by FROOT layer:
| Icon |
Layer |
Plays |
Examples |
|
Foundations |
2 |
AI Landing Zone, Landing Zone Advanced |
|
Reasoning |
3 |
Enterprise RAG, Document Intelligence, AI Search Portal |
|
Orchestration |
8 |
Multi-Agent, Voice AI, IT Tickets, Copilot Teams, Content Moderation |
|
Transformation |
4 |
Fine-Tuning, Model Serving, Prompt Management, Edge AI Phi-4 |
Click any play to see actions:
- Read Documentation opens the play page on frootai.dev
- Init DevKit copies 19+ .github Agentic OS files to your workspace
- Init TuneKit copies AI config + evaluation files
- Init SpecKit generates architecture spec + WAF alignment
- Estimate Cost Azure cost breakdown (dev/prod)
- Run Evaluation auto-runs eval.py + shows quality dashboard
- User Guide opens frootai.dev/user-guide
Title bar:
- Browse opens frootai.dev/solution-plays
Explore all 23 MCP tools across 5 categories:
| Icon |
Category |
Tools |
What They Do |
|
Static |
6 |
Offline knowledge: modules, glossary, architecture patterns |
|
Live |
4 |
Azure docs, external MCP servers, community plays |
|
Agent Chain |
3 |
Build Review Tune workflow |
|
Ecosystem |
4 |
Model catalog, Azure pricing, model comparison |
|
Compute |
6 |
Cost estimation, embeddings, architecture diagrams |
Click any tool opens rich documentation panel with:
- Color-coded badge (green/cyan/yellow/purple/pink per category)
- Full description + input/output docs
- Install commands (npm, pip, Docker)
- Links to frootai.dev + npm + PyPI
Title bar:
- Install setup MCP server (npm/pip/Docker/config)
- Docs opens frootai.dev/setup-guide
- npm opens npmjs.com/package/frootai-mcp
What is a Solution Play?
A Solution Play is a pre-tuned, deployable AI solution like a blueprint you can scaffold in one click.
Each play ships with:
solution-play/
.github/ 19 Agentic OS files
agents/ builder reviewer tuner chain
instructions/ coding standards + patterns
prompts/ /deploy, /test, /review, /evaluate
skills/ deploy-azure, evaluate, tune
hooks/ guardrails.json (policy gates)
config/ AI parameters (temperature, models, guardrails)
evaluation/ test set + eval.py scoring script
infra/ Bicep IaC (main.bicep + parameters.json)
spec/ Architecture spec + WAF alignment
plugin.json Distribution manifest
20 plays available today more coming:
| # |
Play |
What It Deploys |
| 01 |
Enterprise RAG Q&A |
AI Search + OpenAI + Container App |
| 02 |
AI Landing Zone |
VNet + Private Endpoints + RBAC + GPU |
| 03 |
Deterministic Agent |
Reliable agent with guardrails + eval |
| 04 |
Call Center Voice AI |
Real-time speech + sentiment |
| 05 |
IT Ticket Resolution |
Auto-triage + KB resolution |
| 06 |
Document Intelligence |
PDF/image extraction |
| 07 |
Multi-Agent Service |
Orchestrated agent collaboration |
| 08 |
Copilot Studio Bot |
Low-code conversational AI |
| 09 |
AI Search Portal |
Enterprise search with facets |
| 10 |
Content Moderation |
Safety filters + classification |
| 11 |
Landing Zone Advanced |
Multi-region + DR + compliance |
| 12 |
Model Serving AKS |
GPU clusters + endpoints |
| 13 |
Fine-Tuning Workflow |
Data prep train eval deploy |
| 14 |
Cost-Optimized Gateway |
Smart routing + token budgets |
| 15 |
Multi-Modal DocProc |
Images + tables + handwriting |
| 16 |
Copilot Teams Extension |
Teams bot with AI backend |
| 17 |
AI Observability |
Tracing + metrics + alerting |
| 18 |
Prompt Management |
Versioning + A/B testing |
| 19 |
Edge AI Phi-4 |
On-device inference, no cloud |
| 20 |
Anomaly Detection |
Time-series + pattern recognition |
Commands (Ctrl+Shift+P)
| Command |
What |
FrootAI: Browse All Plays |
Quick pick from 20 plays |
FrootAI: Initialize DevKit |
.github Agentic OS (19+ files) |
FrootAI: Initialize TuneKit |
AI config + evaluation files |
FrootAI: Initialize SpecKit |
Architecture spec + WAF alignment |
FrootAI: Initialize Hooks |
guardrails.json |
FrootAI: Initialize Prompts |
4 slash commands |
FrootAI: Setup MCP Server |
npm / pip / Docker / .vscode config |
FrootAI: Quick Cost Estimate |
Azure cost breakdown |
FrootAI: Run Evaluation |
Auto-run eval.py + dashboard |
FrootAI: Auto-Chain Agents |
Build Review Tune workflow |
FrootAI: Validate Config |
Check config/*.json |
FrootAI: Look Up AI Term |
200+ glossary search |
FrootAI: Search Knowledge Base |
Full-text search 16 modules |
FrootAI: Architecture Patterns |
7 decision guides |
FrootAI: Install Plugin |
Community plugin installer |
MCP Server For Your AI Agent
The extension works alongside the FrootAI MCP Server (23 tools that any AI agent can call):
| You (human) |
Your Agent (AI) |
| VS Code Extension |
MCP Server |
| Browse, search, scaffold |
Query knowledge, estimate costs, compare models |
Ctrl+Shift+X FrootAI |
npx frootai-mcp |
Install the MCP Server (click Install in the sidebar, or):
npx frootai-mcp@latest # Node.js
pip install frootai-mcp # Python
docker run -i ghcr.io/frootai/frootai-mcp # Docker
5-Minute Quick Start
- Install
Ctrl+Shift+X Search "FrootAI" Install
- Open sidebar Click the FrootAI tree icon on the left
- Pick a play Click "01 Enterprise RAG Q&A"
- Scaffold Select "Init DevKit" 19 files copied to your workspace
- Add AI config Select "Init TuneKit" config + evaluation copied
- Open Copilot Chat reads your
agent.md + .github/ context automatically
- Deploy Type
/deploy in Copilot Chat guided Azure deployment
The FROOT Framework
Each knowledge module and solution play maps to a FROOT layer:
F Foundations Tokens, models, glossary, Agentic OS
R Reasoning Prompts, RAG, grounding, deterministic AI
O Orchestration Semantic Kernel, agents, MCP, tools
O Operations Azure AI Foundry, GPU infra, Copilot ecosystem
T Transformation Fine-tuning, responsible AI, production patterns
16 modules 200+ terms 60,000+ words of curated content
| Package |
Install |
What |
| Node MCP Server |
npx frootai-mcp |
npm |
| Python SDK |
pip install frootai |
PyPI |
| Python MCP |
pip install frootai-mcp |
PyPI |
| Docker |
docker run -i ghcr.io/frootai/frootai-mcp |
GHCR |
| CLI |
npx frootai <command> |
Docs |
Links
It's simply Frootful.
2026 FrootAI MIT License
