AGI Feeder
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@AGI Feederdecide: Should we expand into the European market?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Billionaire Mindset AGI Feeder v24.2026
An AGI-style task orchestrator that routes tasks to the best AI models, aggregates results with confidence scoring, and applies billionaire thinking frameworks.
What It Does
AGI Feeder is a dual-mode tool: a Model Context Protocol (MCP) server and a CLI that acts as an intelligent AI task orchestrator:
Smart routing — analyzes your task and sends it to the best model (GPT-4o, Claude 3.5, Gemini, DeepSeek R1, etc.)
Multi-model consensus — queries multiple models and synthesizes their responses with confidence scoring
Deep chain-of-thought — 3-stage reasoning: Model A reasons → Model B critiques → Model C synthesizes
Strategic decisions — applies the full Billionaire Mindset framework (First Principles, Risk/Reward, 10x Filter, Anti-fragile)
Billionaire thinking frameworks — 8 frameworks including Pareto 80/20, Asymmetric Risk, Network Effects
Batch orchestration — process arrays of tasks in parallel across models
Related MCP server: LLM Radar
Supported AI Models
Provider | Models |
OpenAI | GPT-4o, GPT-4o Mini, o1, o3-mini |
Anthropic | Claude 3.5 Sonnet, Claude 3.5 Haiku, Claude 3 Opus |
Gemini 2.0 Flash, Gemini 1.5 Pro | |
Mistral | Mistral Large |
DeepSeek | DeepSeek R1, DeepSeek V3 |
All providers are optional — the tool works with any subset of configured API keys.
Installation
# NPM global install
npm install -g billionaire-mindset-agi-feeder
# Or use with npx
npx billionaire-mindset-agi-feeder <command>
# Or clone and build locally
git clone https://github.com/billionaire-mindset/agi-feeder
cd agi-feeder
npm install && npm run buildQuick Start
1. Configure API Keys
# Interactive setup
agi-feeder config
# Or set environment variables
export OPENAI_API_KEY=sk-proj-...
export ANTHROPIC_API_KEY=sk-ant-...
export GOOGLE_API_KEY=AIza...
# Or copy .env.example to .env and fill in
cp .env.example .env2. Route a Task
# Auto-detect task type and route to best model
agi-feeder route "Write a Python function to parse JWT tokens"
# With explicit task type
agi-feeder route "Explain quantum entanglement" --type reasoning
# With specific model
agi-feeder route "Debug this code" --model claude-3-5-sonnet-202410223. Multi-Model Consensus
# Query all configured models, get synthesized answer
agi-feeder consensus "What's the best way to scale a SaaS to $10M ARR?"
# Use specific models
agi-feeder consensus "Rate this business idea" --models gpt-4o,claude-3-5-sonnet-202410224. Deep Chain-of-Thought
# 3-stage reasoning: Reason → Critique → Synthesize
agi-feeder deep "Should I build a B2B or B2C startup?"
# Show full reasoning chain
agi-feeder deep "Solve the trolley problem" --show-chain5. Strategic Decision
# Full billionaire mindset decision analysis
agi-feeder decide "Should I quit my job to start a startup?"
agi-feeder decide "Should we expand into the European market?"6. Apply Mindset Frameworks
# List all frameworks
agi-feeder mindset list
# First Principles analysis
agi-feeder mindset first-principles "Why is my startup not growing?"
# Pareto 80/20
agi-feeder mindset pareto-8020 "How should I allocate my time as a founder?"
# Asymmetric Risk
agi-feeder mindset asymmetric-risk "Should I invest $50k in this opportunity?"
# 10x Moonshot
agi-feeder mindset 10x-moonshot "How do I 10x my consulting revenue?"
# Anti-fragile Strategy
agi-feeder mindset anti-fragile "How do I make my startup resilient to market crashes?"
# Infinite Leverage
agi-feeder mindset infinite-leverage "What's the highest ROI way to grow my audience?"
# Time Billionaire
agi-feeder mindset time-billionaire "Should I hire a VA or do these tasks myself?"
# Network Effects
agi-feeder mindset network-effects "How do I build a marketplace with strong network effects?"7. Batch Processing
# Create a batch file
cat > tasks.json << 'EOF'
{
"tasks": [
{ "id": "1", "task": "Summarize quantum computing in 3 sentences", "type": "analysis" },
{ "id": "2", "task": "Write a Python hello world", "type": "coding" },
{ "id": "3", "task": "Name 5 ways to generate passive income", "type": "creative" }
]
}
EOF
agi-feeder batch tasks.json
# Or use the example file
agi-feeder batch examples/batch-tasks.json8. Check Model Status
agi-feeder status
agi-feeder status --provider openaiMCP Server
AGI Feeder runs as a Model Context Protocol server, allowing AI assistants (Claude Desktop, Cursor, etc.) to use it as a tool.
Start the Server
# Via CLI
agi-feeder serve
# Or directly
node bin/agi-feeder.js serveClaude Desktop Integration
Add to your claude_desktop_config.json:
{
"mcpServers": {
"agi-feeder": {
"command": "node",
"args": ["/path/to/agi-feeder/bin/agi-feeder.js", "serve"],
"env": {
"OPENAI_API_KEY": "sk-...",
"ANTHROPIC_API_KEY": "sk-ant-...",
"GOOGLE_API_KEY": "AIza..."
}
}
}
}Available MCP Tools
Tool | Description |
| Route task to best model with auto type detection |
| Multi-model consensus with confidence scoring |
| 3-stage deep reasoning chain |
| Billionaire mindset decision framework |
| Parallel batch task processing |
| Check model availability and API key status |
| Apply any of 8 billionaire thinking frameworks |
Billionaire Mindset Frameworks
Framework | Tagline |
| Decompose to bedrock truth, rebuild from scratch |
| 20% inputs drive 80% of outputs |
| Limited downside, 100x upside bets |
| 10x is easier than 2x — abandon incrementalism |
| Systems that get stronger from stress |
| Code, media, capital — zero marginal cost leverage |
| Evaluate by time-freedom generated |
| Score by flywheel and network potential |
Programmatic API
import {
routeTask,
multiModelConsensus,
chainOfThoughtDeep,
applyMindsetFramework,
strategicDecision,
} from 'billionaire-mindset-agi-feeder';
// Route a task
const result = await routeTask({
task: 'Build a REST API in Node.js',
taskType: 'coding',
maxTokens: 2048,
});
console.log(result.selectedModel); // e.g. "claude-3-5-sonnet-20241022"
console.log(result.result.response);
// Multi-model consensus
const consensus = await multiModelConsensus({ task: 'What is the future of AI?' });
console.log(consensus.aggregated.synthesis);
console.log(`Agreement: ${consensus.aggregated.agreement * 100}%`);
// Deep chain of thought
const deep = await chainOfThoughtDeep({ task: 'Is consciousness emergent?' });
console.log(deep.finalSynthesis);
// Apply a mindset framework
const mindset = await applyMindsetFramework('first-principles', 'Why is user acquisition expensive?');
console.log(mindset.insights);
mindset.actionItems.forEach((a) => console.log(`${a.action} (${a.impact})`));
// Strategic decision
const decision = await strategicDecision('Should I raise VC funding?');
console.log(decision.analysis.recommendation);Configuration
Environment Variables
Variable | Default | Description |
| — | OpenAI API key |
| — | Anthropic API key |
| — | Google/Gemini API key |
| — | Mistral API key |
| — | DeepSeek API key |
|
| Default model (or "auto" for routing) |
|
| Output format: text/json/markdown |
|
| Max API retry attempts |
|
| Request timeout (ms) |
|
| Log level: debug/info/warn/error |
Config File
Saved to ~/.agi-feeder/config.json. Managed via agi-feeder config.
Docker
# Build and run
docker build -t agi-feeder .
docker run -e OPENAI_API_KEY=sk-... -e ANTHROPIC_API_KEY=sk-ant-... agi-feeder
# Docker Compose
cp .env.example .env # fill in your keys
docker-compose up
# CLI via Docker
docker run --rm -e OPENAI_API_KEY=sk-... agi-feeder \
node bin/agi-feeder.js route "Write a Fibonacci function in Rust"Railway Deployment
railway login
railway new
railway up
railway variables set OPENAI_API_KEY=sk-... ANTHROPIC_API_KEY=sk-ant-...Development
git clone https://github.com/billionaire-mindset/agi-feeder
cd agi-feeder
npm install
npm run build
# Run CLI without building (requires ts-node)
npm install ts-node --save-dev
npx ts-node src/cli/index.ts status
# Watch mode
npm run build:watchArchitecture
src/
├── types/ — TypeScript type definitions
├── core/
│ ├── config.ts — Configuration management
│ ├── models.ts — Model registry + API calls (OpenAI, Anthropic, Google, Mistral, DeepSeek)
│ ├── router.ts — Intelligent task router with type detection
│ ├── aggregator.ts — Multi-model consensus + chain-of-thought + batch
│ └── mindset.ts — 8 billionaire thinking frameworks engine
├── mcp-server/
│ ├── tools.ts — MCP tool definitions and handlers
│ └── index.ts — MCP server (stdio transport)
└── cli/
├── index.ts — Commander.js CLI commands
└── interactive.ts — Inquirer.js interactive prompts + formattersLicense
MIT — Built by Billionaire Mindset AGI
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP server for building and testing AI agents with multi-model experimentation and insights.
MCP server unifying ERPs, CRMs, APIs and knowledge base for Claude, ChatGPT and Gemini.
MCP server for VC pitch-deck scoring, thesis-fit matching, and deal-flow management.
MCP server connecting AI agents to non-custodial staking data across 130+ networks.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceAn MCP server that enables multi-model debate and consensus building through a single tool. It orchestrates multiple AI models from various providers to debate topics and reach validated conclusions with real-time progress tracking.8 npm3MIT
- AlicenseAqualityFmaintenanceMCP server that provides AI assistants with real-time, current information about AI models from OpenAI, Anthropic, and Google, including pricing, capabilities, and context windows.4MIT
- AlicenseNot gradedqualityBmaintenanceMCP server that enables AI agents to run a deterministic orchestration loop with decomposition, subagent execution, and review feedback across multiple LLM backends.60MIT
- AlicenseNot gradedqualityDmaintenanceAn MCP server that orchestrates multiple AI agents in parallel to get diverse perspectives on a single topic, supporting debate, review, and quick modes.77 npmApache 2.0