pich/ai-economics-mcp

Overview

MCP server offering 12 calculators for AI economics, including token cost, LLM energy, agent-hour, Proof-Adjusted Autonomy, and revocation exposure. It provides a free API without requiring an API key. Built by Michał Piszczek in JavaScript.

Best for

Best for
Developers needing AI cost and energy calculations within MCP-based workflows.

Use cases

• Estimate token costs for LLM API usage
• Calculate energy consumption of AI models
• Assess agent-hour economics and autonomy metrics

How to use

Install

claude mcp add ai-economics -- npx -y @michalpiszczek/ai-economics-mcp

Tools exposed

• token_cost
• context_window
• agent_hour
• model_routing
• llm_energy
• joules_per_verified_task
• token_burn
• humanoid_energy
• proof_adjusted_autonomy
• revocation_exposure
• verification_bottleneck
• proof_debt
• AI_ECONOMICS_API

Tested with

Claude Desktop, Claude Code, Cursor

Example client config

{\n  "mcpServers": {\n    "ai-economics": {\n      "command": "npx",\n      "args": ["-y", "@michalpiszczek/ai-economics-mcp"]\n    }\n  }\n}

Notes

MCP server offering 12 calculators for AI economics, including token cost, LLM energy, agent-hour, Proof-Adjusted Autonomy, and revocation exposure. It provides a free API without requiring an API key. Built by Michał Piszczek in JavaScript.

1 stars on GitHub. Last updated 2026-09-10. Licensed MIT.

Use cases

• Estimate token costs for LLM API usage
• Calculate energy consumption of AI models
• Assess agent-hour economics and autonomy metrics

Pros

• Free to use with no API key required
• Covers multiple economic metrics in one server
• Integrates with MCP clients for easy access

Cons

• Low community adoption (1 star on GitHub)
• Limited documentation or examples may be available
• Requires MCP client setup to use

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.