System Online

Structured reasoning
for AI agents

120 validated mental models your agents can call via API. No signup. No complexity. Just working infrastructure for building agents that think.

Explore 120 Models → Book Discovery Call

api.hummbl.workers.dev

$ curl https://hummbl-api.hummbl.workers.dev/v1/models/P1

120 Mental Models

6 API Endpoints

5 Vendor Runtimes

66 AI Agents

-- API Latency

The Framework

Base120 organizes cognitive tools across 6 transformation types. Each provides a different lens for problem-solving.

Perspective

Frame and anchor viewpoint. Change how you see the problem.

Inversion

Flip to see what avoiding failure reveals. Think backwards.

Composition

Combine for emergent properties. Build complexity from simplicity.

Decomposition

Break complexity down. Understand the parts.

Recursion

Apply patterns at multiple scales. Iterate and refine.

Systems

Understand governing rules. See the meta-patterns.

Try It Now

Describe a problem you're facing and get recommended mental models instantly. No signup required.

Recommended Models

Explore All 120 Models →

Claude Desktop Setup

Install the MCP server to give Claude direct access to all 120 models.

Step 1: Install

terminal

$ npm install -g @hummbl/mcp-server

Step 2: Configure

Add to claude_desktop_config.json

claude_desktop_config.json

{
  "mcpServers": {
    "hummbl": {
      "command": "npx",
      "args": ["@hummbl/mcp-server"]
    }
  }
}

Step 3: Available Tools

select_model Select appropriate mental models for a problem
apply_transformation Apply a transformation type to reframe a problem (P, IN, CO, DE, RE, SY)
analyze_problem Comprehensive analysis using multiple mental models

Step 4: Available Resources

models://all All 120 mental models
models://by-transformation Models grouped by transformation type
relationships://all Validated relationships between mental models
transformations://overview Overview of the 6 transformations

Step 5: Available Prompts

problem_decomposition Systematic 6-transformation problem analysis framework

Use Cases

HUMMBL provides structured reasoning infrastructure for AI systems. Use it via API, Claude Desktop MCP integration, or direct framework integration:

Agent Frameworks

LangChain, CrewAI, AutoGPT, Semantic Kernel - invoke mental models programmatically to guide agent reasoning.

Claude Desktop

MCP server gives Claude direct access to all 120 models. Search, retrieve, and apply cognitive frameworks in real-time.

Custom Applications

REST API for decision support tools, prompt engineering systems, or reasoning-enhanced workflows. No framework required.

Prompt Engineering

Retrieve relevant mental models based on problem context, then inject into system prompts for better LLM outputs.

Knowledge Systems

Build cognitive enhancement tools, decision support systems, or structured thinking applications on top of Base120.

Research & Education

Access validated mental models programmatically for cognitive science research, educational tools, or training systems.

Book a call to see how Base120 fits your use case.

Schedule Discovery Call →