Skip to main content

FrootAI — AmpliFAI your AI Ecosystem Get Started

npm

npm — FAI MCP Server

Add AI Infra, Platform, and App Knowledge to Your Agent

Quick Run
npx frootai-mcp
Install Global
npm i -g frootai-mcp
Hosted MCP (v1.0)
https://mcp.frootai.dev/mcp

Federation Router — one connection, every Tier-1 area

FrootAI MCP v6 stops being one tool surface and becomes a federation router: one MCP connection that exposes every Tier-1 tool from every Tier-1 publisher, namespaced as <area>.<tool>, gated by a single trust manifest you control. Stdio for local-first, or hosted at mcp.frootai.dev for centrally-managed teams.

Enterprise-ready Every attach passes the MCP Trust Gate — 4 trust tiers, byte-identical 3-way mirror, no silent-attach path.

Federated MCP Areas — 21 first-party + verified-publisher servers

Each area is a published MCP server the federation router attaches on demand. Sourced from mcp-servers-seed.json; mirrored to /data/mcp-marketplace.json + /data/mcp-trust.json.

azure
MS
Azure MCP
Federated access to Azure ARM, ACR, AKS, App Service, and cost-management APIs.
publisher: microsoft
github
MS
GitHub MCP
Federated access to GitHub repos, PRs, issues, and Actions.
publisher: github
playwright
MS
Playwright MCP
Headless-browser automation: navigate, click, type, screenshot.
publisher: microsoft
markitdown
MS
MarkItDown MCP
Document-to-markdown conversion (.docx, .pdf, .xlsx, etc.).
publisher: microsoft
ms-learn
MS
MS Learn MCP
Search + retrieve content from learn.microsoft.com.
publisher: microsoftdocs
context7
VERIFIED
Context7 MCP
Up-to-date library docs grounding for code-generation copilots.
publisher: upstash
chrome-devtools
VERIFIED
Chrome DevTools MCP
Chrome DevTools Protocol surface for browser introspection + automation.
publisher: chromedevtools
ms-fabric
MS
Microsoft Fabric MCP
Microsoft Fabric data + analytics surface (Power BI, Synapse, OneLake).
publisher: microsoft
azure-devops
MS
Azure DevOps MCP
Enterprise SCM + boards + pipelines — complement to GitHub for Azure-hosted teams.
publisher: microsoft
ms-nuget
MS
Microsoft NuGet MCP
NuGet package intelligence for .NET-track Plays — search, version, dependency tree.
publisher: nuget
awesome-copilot
MS
Awesome Copilot MCP
Browse Copilot customisations from the Awesome Copilot repo.
publisher: microsoft
notion
VERIFIED
Notion MCP
Notion pages + databases + comments — where specs live for many teams.
publisher: makenotion
terraform
VERIFIED
Terraform MCP
Terraform module discovery + provider docs — IaC complement to Azure / Bicep.
publisher: hashicorp
firecrawl
VERIFIED
Firecrawl MCP
Best-in-class web data extraction — seed grounding from arbitrary websites.
publisher: firecrawl
tavily
VERIFIED
Tavily MCP
Advanced web search optimised for AI agents — the agentic retriever persona.
publisher: tavily-ai
supabase
VERIFIED
Supabase MCP
Postgres + auth + storage + edge functions — for app-builder Plays.
publisher: supabase
stripe
VERIFIED
Stripe MCP
Customers, subscriptions, invoices, products — for SaaS / billing scaffolds.
publisher: stripe
mongodb
VERIFIED
MongoDB MCP
MongoDB Atlas + self-hosted clusters — for NoSQL Plays.
publisher: mongodb-js
atlassian
VERIFIED
Atlassian MCP
Jira issues + Confluence pages + Rovo — for enterprise PM Plays.
publisher: atlassian
vercel
VERIFIED
Vercel MCP
Next.js project inspection + deploy preview — pairs with the Next.js Orchard Plays.
publisher: vercel
sentry
VERIFIED
Sentry MCP
Issue search + alert rules + release tracking — for any prod-Play.
publisher: getsentry

Federation Lifecycle Tools — the 6 fai_* primitives

These are NOT the in-process tools below — they orchestrate attach/detach of the 21 federated areas above. Wired in M2; expanded across M3-M9. Every agent / skill / play / plugin can call these directly via the MCP protocol. Browse the full 100-server registry at /mcp-tooling/marketplace.

fai_attach_mcp
Attach a federated MCP area; runs trust gate, spawns stdio/HTTP transport, namespaces tools as <area>.<tool>
fai_detach_mcp
Detach + dispose an attached area; idle disconnect honored at the configured timeout
fai_list_attached
List all currently-attached areas with sessionId, attachedAt, lastUsedAt, tool count
fai_discover_mcp
Discover available MCP areas from the seed roster + community catalog (raw-marketplace.json)
fai_trust_query
Query the trust manifest for a publisher: returns tier (first-party-ms / verified / community / untrusted) + policy
fai_invoke_via
One-shot invoke: attach → invoke <area>.<tool>(args) → detach in a single call (for ephemeral usage)

Trust Manifest — 4 tiers, gated at attach

Every fai_attach_mcp passes through this gate. There is no bypass. Published at /data/mcp-trust.json; same shape across npm-mcp + python-mcp + web (M0.12 drift-protected).

first-party-ms
auto-attachdestructive ok
Microsoft + GitHub-published areas. Attach without prompt; destructive tool calls allowed.
verified-publisher
auto-attachread-only
Curated publishers (OpenAI, HashiCorp, Upstash, Notion, Stripe, Tavily, etc.). Attach without prompt; destructive calls require explicit confirmation.
community
promptread-only
Community-published areas. Trust prompt at attach; no destructive calls.
untrusted
promptread-only
Explicitly blocklisted. Attach is rejected at the gate; never reaches transport spawn.

Without FrootAI MCP

  • Agent searches the internet — slow, noisy
  • Burns 5,000+ tokens per architecture query
  • May hallucinate design guidance
  • Generic answers — no Azure patterns
FrootAI

With FrootAI MCP

  • Queries curated 664KB knowledge base
  • 90% less token burn
  • Zero hallucination — grounded in verified docs
  • Azure-specific best practices & patterns

Connect to Your Client

Claude Desktop
claude_desktop_config.json
{"mcpServers":{"frootai":{"command":"npx","args":["frootai-mcp"]}}}
VS Code / Copilot
.vscode/mcp.json
{"servers":{"frootai":{"command":"npx","args":["frootai-mcp"]}}}
Microsoft Foundry
Agent → Tools → Add MCP
Point to npx frootai-mcp
Cursor
Cursor
MCP settings
{"mcpServers":{"frootai":{"command":"npx","args":["frootai-mcp"]}}}
Windsurf
MCP settings
{"mcpServers":{"frootai":{"command":"npx","args":["frootai-mcp"]}}}
Copilot Studio
Copilot Studio tools
Add MCP connector
Gemini / Codex
MCP config
{"mcpServers":{"frootai":{"command":"npx","args":["frootai-mcp"]}}}

Tools Agent Receives

Static + live + agent chain + ecosystem + FAI engine + scaffold + marketplace = full MCP toolkit

list_modules
Browse knowledge modules by FROOT layer
get_module
Read any module content (F1–T3)
lookup_term
Comprehensive AI/ML term definitions
search_knowledge
Full-text search all modules
get_architecture_pattern
7 pre-built decision guides
get_froot_overview
Complete FROOT framework summary

Live Tools (v2 — network-enabled)

fetch_azure_docs
Search Microsoft Learn for Azure docs
fetch_external_mcp
Find MCP servers from registries
list_community_plays
List solution plays from GitHub
get_github_agentic_os
.github 7 primitives guide

Agent Chain Tools (Build → Review → Tune)

agent_build
Play-aware architecture, code patterns, pinned deps (v6.3+)
agent_review
12-rule code engine: severity + file:line + suggestion (v6.3+)
agent_tune
Validates openai.json/guardrails/play configs against thresholds (v6.3+)

AI Ecosystem Tools (v3)

get_model_catalog
Compare GPT-4o, Claude, Llama, Phi models
get_azure_pricing
Token pricing for Azure OpenAI models
estimate_cost
Estimate monthly cost for a solution play
compare_models
Side-by-side model comparison
compare_plays
Compare two solution plays
semantic_search_plays
Natural language play search
embedding_playground
Compare texts for semantic similarity — v6.7+ real Azure OpenAI 1536-dim embeddings (auto when AZURE_OPENAI_ENDPOINT set) or char-ngram cosine fallback

Developer Tools

validate_config
Validate TuneKit configs (openai.json, guardrails.json) against best practices
generate_architecture_diagram
Generate Mermaid.js architecture diagram for any solution play
run_evaluation
Quality scoring — groundedness, relevance, coherence, fluency (pass/fail)

FAI Engine — The UniFAIng Glue (FAI Protocol Runtime)

wire_play
Resolve FAI Protocol context, bind primitives, activate guardrails
inspect_wiring
Audit a play's wiring — primitives, context, health score
validate_manifest
Validate fai-manifest.json schema + completeness
get_play_detail
Full play spec — services, config, WAF alignment, evaluation
list_primitives
Browse primitives by type (agents, skills, hooks…)
evaluate_quality
Run quality evaluation with configurable pass/fail thresholds
run_eval_live
POST cases to a live endpoint, score with simple/LLM-as-judge, aggregate mean/p95 vs threshold (v6.6+)

Scaffold & Create — Generate Production-Ready Plays

scaffold_play
Generate a complete play (24+ files) with FAI Protocol auto-wired
create_primitive
Create an agent, instruction, or skill with proper frontmatter
smart_scaffold
Semantic search → find best matching play → scaffold it
scaffold_component
Copy-paste-ready code for a single Azure AI building block (15 components, py/ts/cs) (v6.4+)
get_play_config
Starter configs for a play: openai/guardrails + play-specific (search.json, speech.json, etc.) (v6.4+)
get_dependencies
Pinned dependency lists per language (python/typescript/csharp) for a play (v6.4+)
generate_bicep
Play-specific Bicep IaC with Managed Identity, RBAC, tags, optional private endpoints (v6.5+)

Workspace — Detect, Inspect, Recommend

analyze_workspace
Inspect cwd — detected play, primitives, configs, Bicep services, FAI manifest, gaps + recommendations

Marketplace — npm for AI Primitives (77+ Plugins)

marketplace_search
Semantic search across plugins by use case
marketplace_browse
Paginated listing with 8 category filters
install_plugin
Copy plugin primitives into .github/ with conflict detection
uninstall_plugin
Remove plugin primitives cleanly
list_installed
Scan .github/ and match against marketplace registry
check_compatibility
Validate plugin + play alignment + WAF + file conflicts
validate_plugin
Check plugin.json schema, naming, and file refs
compose_plugins
Multi-install with cross-plugin conflict detection
publish_plugin
Validate → generate marketplace entry → register
check_plugin_updates
Compare installed vs marketplace versions
resolve_dependencies
Topological dependency resolution + install order
list_external_plugins
Browse community plugins from external sources
marketplace_stats
Full analytics: totals, categories, top plugins, play coverage

Orchard MCP Server — Harvest Pipeline Tools

8 tools that drive the harvest pipeline: discover repos, extract architecture facts, find exemplars, import + customize Solution Plays. Paid tools require an active subscription.

discover_repo
FREE

GitHub metadata + license + commit SHA — deterministic, no LLM

60/min
Invocation
discover_repo({ url: "https://github.com/Azure/azure-search-openai-demo" })
Response shape ▸
{
  "owner": "Azure",
  "name": "azure-search-openai-demo",
  "license": "MIT",
  "commit_sha": "abc123def456",
  "default_branch": "main",
  "stars": 5200,
  "language": "Python",
  "topics": ["azure", "openai", "rag"]
}
extract_facts
FREE

Shallow clone → RepoFacts JSON — architecture, services, languages

20/min
Invocation
extract_facts({ url: "https://github.com/Azure/azure-search-openai-demo" })
Response shape ▸
{
  "services": ["Azure AI Search", "Azure OpenAI", "App Service"],
  "languages": { "Python": 72, "Bicep": 18, "TypeScript": 10 },
  "iac_tool": "bicep",
  "has_docker": true,
  "entry_points": ["app/backend/app.py"]
}
retrieve_exemplars
FREE

Top-3 reference plays via embeddings for scaffold grounding

30/min
Invocation
retrieve_exemplars({ query: "RAG with Azure AI Search" })
Response shape ▸
[
  { "slug": "enterprise-rag", "similarity": 0.95 },
  { "slug": "contoso-chat", "similarity": 0.82 },
  { "slug": "aks-store-demo", "similarity": 0.71 }
]
list_plays
FREE

Browse all harvested Solution Plays with filter + sort

60/min
Invocation
list_plays({ source: "harvested", sort: "confidence" })
Response shape ▸
{
  "total": 142,
  "plays": [
    { "slug": "enterprise-rag", "confidence": 0.95, "source": "harvested" },
    ...
  ]
}
import_accelerator
PAID

Harvest → package → sign → deliver a Solution Play ZIP

5/min
Invocation
import_accelerator({ url: "https://github.com/Azure/azure-search-openai-demo" })
Response shape ▸
{
  "run_id": "a1b2c3d4",
  "status": "ready",
  "zip_url": "https://cdn.frootai.dev/plays/...",
  "manifest_url": "https://cdn.frootai.dev/plays/.../manifest.json",
  "cosign_verify": "cosign verify-blob ..."
}
Auth required: Requires OAuth device-flow auth. Run npx frootai-mcp --login to authenticate, then invoke. Unauthenticated calls return a device_code + verification_uri for browser-based sign-in.
Upgrade to unlock →
customize_play
PAID

Apply org policy defaults (regions, RBAC, naming) before import

10/min
Invocation
customize_play({ slug: "enterprise-rag", regions: ["westeurope"] })
Response shape ▸
{
  "slug": "enterprise-rag",
  "applied": { "regions": ["westeurope"], "naming": "contoso-{service}" },
  "diff_url": "https://frootai.dev/.../customize/preview"
}
Auth required: Requires active paid subscription. Authenticate via npx frootai-mcp --login first.
Upgrade to unlock →
re_harvest
PAID

Re-run the harvest pipeline on an existing play (freshness)

5/min
Invocation
re_harvest({ slug: "enterprise-rag" })
Response shape ▸
{
  "run_id": "e5f6g7h8",
  "status": "running",
  "previous_sha": "abc123",
  "new_sha": "def456"
}
Auth required: Requires active paid subscription. Authenticate via npx frootai-mcp --login first.
Upgrade to unlock →
approve_play
FOUNDER

Founder-only: approve a harvested play for the public catalog

10/min
Invocation
approve_play({ slug: "enterprise-rag", tier: "gold" })
Response shape ▸
{
  "slug": "enterprise-rag",
  "approved": true,
  "tier": "gold",
  "published_at": "2026-06-14T12:00:00Z"
}
Auth required: Founder-only tool. Requires authenticated founder session — not available to regular subscribers.

Verified MCP Server — AVM Intelligence Tools

5 tools powered by Azure Verified Modules: browse the AVM catalog, validate WAF/CAF compliance, and compute cost + carbon footprint for composed infrastructure.

list_avm_modules
FREE

Query the AVM catalog with provider/type/tag filters

60/min
Invocation
list_avm_modules({ provider: "Microsoft.Network" })
Response shape ▸
{
  "total": 47,
  "modules": [
    { "name": "res/network/virtual-network", "version": "0.5.1", "provider": "Microsoft.Network" },
    { "name": "res/network/private-endpoint", "version": "0.3.0", "provider": "Microsoft.Network" },
    ...
  ]
}
get_avm_module
FREE

Fetch cached module content, docs, and version history

60/min
Invocation
get_avm_module({ name: "res/network/virtual-network" })
Response shape ▸
{
  "name": "res/network/virtual-network",
  "version": "0.5.1",
  "description": "Azure Virtual Network with subnets, NSGs, and peering",
  "docs_url": "https://github.com/Azure/bicep-registry-modules/...",
  "versions": ["0.5.1", "0.4.0", "0.3.2"],
  "parameters": ["name", "addressPrefixes", "subnets", "tags"]
}
validate_waf_compliance
PAID

Validate composed Bicep/Terraform against WAF + CAF gates

10/min
Invocation
validate_waf_compliance({ bicepPath: "infra/main.bicep" })
Response shape ▸
{
  "compliant": true,
  "score": 92.5,
  "pillars": {
    "reliability": { "score": 95, "findings": [] },
    "security": { "score": 88, "findings": ["Enable private endpoints"] },
    "cost": { "score": 94, "findings": [] },
    "operational": { "score": 93, "findings": [] },
    "performance": { "score": 92, "findings": [] }
  }
}
Auth required: Requires active paid subscription. Authenticate via npx frootai-mcp --login first.
Upgrade to unlock →
compute_cost
PAID

Estimate monthly Azure cost for composed infrastructure

20/min
Invocation
compute_cost({ slug: "enterprise-rag", region: "westeurope" })
Response shape ▸
{
  "slug": "enterprise-rag",
  "region": "westeurope",
  "monthly_eur": 147.50,
  "band": "€100–200/mo",
  "breakdown": [
    { "service": "Azure AI Search", "sku": "S1", "eur": 72.00 },
    { "service": "Azure OpenAI", "sku": "gpt-4o", "eur": 45.00 },
    { "service": "App Service", "sku": "P1v3", "eur": 30.50 }
  ],
  "priced_at": "2026-06-14"
}
Auth required: Requires active paid subscription. Authenticate via npx frootai-mcp --login first.
Upgrade to unlock →
compute_carbon
PAID

Estimate g CO₂e/mo carbon footprint for composed infra

20/min
Invocation
compute_carbon({ slug: "enterprise-rag", region: "westeurope" })
Response shape ▸
{
  "slug": "enterprise-rag",
  "region": "westeurope",
  "g_co2e_month": 1240,
  "band": "Low",
  "breakdown": [
    { "service": "Azure AI Search", "g_co2e": 520 },
    { "service": "Azure OpenAI", "g_co2e": 480 },
    { "service": "App Service", "g_co2e": 240 }
  ],
  "methodology": "Azure Carbon Optimization API"
}
Auth required: Requires active paid subscription. Authenticate via npx frootai-mcp --login first.
Upgrade to unlock →

Install Guide — All 3 MCP Servers

Add the main FAI MCP server + the Orchard harvest server + the Verified AVM server to your AI agent in one config.

VS Code / GitHub Copilot.vscode/mcp.json
{
  "servers": {
    "frootai": { "command": "npx", "args": ["frootai-mcp"] },
    "frootai-harvest": { "command": "npx", "args": ["frootai-mcp", "--server", "harvest"] },
    "frootai-verified": { "command": "npx", "args": ["frootai-mcp", "--server", "verified"] }
  }
}
CursorSettings → MCP → Add Server
{
  "mcpServers": {
    "frootai": { "command": "npx", "args": ["frootai-mcp"] },
    "frootai-harvest": { "command": "npx", "args": ["frootai-mcp", "--server", "harvest"] },
    "frootai-verified": { "command": "npx", "args": ["frootai-mcp", "--server", "verified"] }
  }
}
Claude Desktopclaude_desktop_config.json
{
  "mcpServers": {
    "frootai": { "command": "npx", "args": ["frootai-mcp"] },
    "frootai-harvest": { "command": "npx", "args": ["frootai-mcp", "--server", "harvest"] },
    "frootai-verified": { "command": "npx", "args": ["frootai-mcp", "--server", "verified"] }
  }
}
Continue.dev~/.continue/config.json → mcpServers
{
  "mcpServers": [
    { "name": "frootai", "command": "npx", "args": ["frootai-mcp"] },
    { "name": "frootai-harvest", "command": "npx", "args": ["frootai-mcp", "--server", "harvest"] },
    { "name": "frootai-verified", "command": "npx", "args": ["frootai-mcp", "--server", "verified"] }
  ]
}
Zedsettings.json → language_servers → mcp
{
  "language_servers": {
    "mcp": {
      "servers": {
        "frootai": { "command": "npx", "args": ["frootai-mcp"] },
        "frootai-harvest": { "command": "npx", "args": ["frootai-mcp", "--server", "harvest"] },
        "frootai-verified": { "command": "npx", "args": ["frootai-mcp", "--server", "verified"] }
      }
    }
  }
}

See It In Action

Real conversations with the FAI MCP server — what you ask and what you get back

🔍
You ask: "I want to build a RAG chatbot"
semantic_search_plays({ query: "RAG chatbot with Azure AI Search" })
Result: Play 01 (Enterprise RAG, 92% match), Play 21 (Agentic RAG, 78%), Play 09 (AI Search Portal, 65%)
Instead of Googling, your AI agent instantly finds the best pre-built solution from all plays — ranked by BM25 relevance.
🏗️
You ask: "Scaffold a new project from scratch"
scaffold_play({ name: "Invoice Processor", model: "gpt-4o" })
Result: 24 files created — agents, skills, config, infra, evaluation — all auto-wired via FAI Protocol
One tool call creates a complete, production-ready project. No other MCP server does this.
🔌
You ask: "Wire my play and check if everything connects"
wire_play({ playId: "01" })
Result: ✅ All primitives wired: 1 agent → 3 instructions → 3 skills → 3 hooks, shared context (4 knowledge modules + 5 WAF pillars), 43ms
"Wiring" = the FAI Engine reads your manifest, resolves all paths, injects shared context, and verifies quality gates. Like docker compose up for AI primitives.
📊
You ask: "Check if my AI responses meet quality standards"
evaluate_quality({ scores: { groundedness: 0.97, relevance: 0.88, safety: 0 } })
Result: ✅ PASS — Groundedness 0.97 ≥ 0.95, Relevance 0.88 ≥ 0.85, Safety 0 = 0. Each play has its own quality bar.
Guardrails are defined per-play in fai-manifest.json. Fail a threshold → the engine tells you which metric to fix.
You ask: "Build → Review → Tune workflow"
agent_build → agent_review → agent_tune("Invoice processing API with Azure Document Intelligence")
Result: Build: architecture guidance → Review: security + quality audit → Tune: production readiness verdict
Mirrors how real teams work: architect → code reviewer → QA. Three agents forming a quality chain.

What Makes This Different

Other MCP servers wrap APIs. FrootAI MCP is a domain knowledge engine with a protocol that binds AI primitives together.

FeatureGitHub MCPAnthropic MCPFAI MCP
Runtime engineNoNo✅ FAI Engine
Wires primitivesNoNo✅ fai-manifest.json
Quality gatesNoNo✅ Guardrails per play
Scaffolds projectsNoNo✅ 24 files, auto-wired
Knowledge baseNoNo✅ Modules, BM25
Plugin marketplaceNoNo✅ 77+ plugins
Progress tokensNoNo✅ Per-file progress
Tool annotationsPartialNo✅ All tools