Skip to main content
Glama
Sign In
ai-economics-mcp

ai-economics-mcp

by pich
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Remote

ai-economics-mcp

DOI

npm Glama score MIT

MCP server for AI economics. Gives Claude, Cursor and any MCP client 12 calculators for the questions that decide AI budgets: what tokens cost, what queries burn, and whether agent work can be proven.

Listed in the official MCP Registry as pl.piszczek/ai-economics (domain-verified). Wraps the free piszczek.pl/tools API — no key, no sign-up, stateless (inputs are never stored). The math is the same as the interactive calculators.

Install

Claude Desktop / Claude Code (claude_desktop_config.json / .mcp.json):

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

Claude Code one-liner:

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

Related MCP server: x402 MCP Server

Tools

Tool

Answers

token_cost

What do 1M tokens cost per month across GPT, Claude, Gemini, DeepSeek?

context_window

How many tokens is 50 pages, does it fit, what does carrying it cost?

agent_hour

What does one hour of an AI agent cost, fully loaded with verification?

model_routing

How much does routing to a cheaper tier save per year?

llm_energy

How much electricity does an AI query use?

joules_per_verified_task

Which model is most efficient per task that actually passes?

token_burn

What does org-wide token burn cost in money, kWh, CO₂?

humanoid_energy

How long can a humanoid robot run per charge?

proof_adjusted_autonomy

How autonomous is a "90% autonomous" agent once proof is required?

revocation_exposure

How long does a revoked token keep working across your stack?

verification_bottleneck

How many AI agents can a human review team absorb?

proof_debt

What does unverified AI work cost over time?

All parameters are optional; defaults mirror the interactive calculators. Every response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Example

You: how autonomous is our agent really? it completes 90% alone, evidence for 95%, 80% independently validated, 90% on time

Claude (calls proof_adjusted_autonomy): PAA = 61.6% — supervised autonomy. The gap vs the claimed 90% is 28.4 pp, and the biggest lever is independent validation…

Configuration

Env var

Default

Purpose

AI_ECONOMICS_API

https://piszczek.pl/tools/api

Point at a self-hosted instance

Glama builds and releases

Glama generates its own Dockerfile from the build configuration. Use these settings:

  • Build steps: ["npm ci --omit=dev"]

  • CMD arguments: ["mcp-proxy", "--", "node", "index.js"]

  • Environment variables: optional AI_ECONOMICS_API; no credentials required.

This is plain JavaScript: there is no build script or compilation step. Do not use pnpm run build or npm run build. After syncing the repository, build the selected commit and publish a Glama release from the successful test. A GitHub or npm release alone does not trigger Glama's quality evaluation.

For local stdio use, the repository Dockerfile needs no proxy or exposed ports:

docker build -t ai-economics-mcp .
docker run -i --rm ai-economics-mcp

Concepts behind the tools

Citing a result

Every tool response carries a cite_as sentence written to be quoted verbatim. For a document that has to survive review, DOIs and BibTeX for this server, the calculators and the three concepts they instrument are at piszczek.pl/cite.

License

MIT. Concepts and calculators by Michał Piszczek (CC BY 4.0 — attribution appreciated).

Available Tools

12 tools
agent_hourA

Agent-Hour Cost: Fully-loaded cost of one AI agent-hour: compute plus human verification, vs the human hour it replaces. All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/agent-hour. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: price, tokens_m, human_rate, review_min.

context_windowA

Context Window: How many tokens is this content, does it fit the window, and what does carrying it cost per request? All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/context-window. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: unit, price, amount, window.

humanoid_energyA

Humanoid Energy Budget: How long can a humanoid robot run per charge? Splits the battery between actuation, inference and idle. All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/humanoid-energy. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: duty, idle_w, compute_w, actuation_w, +1 more.

joules_per_verified_taskA

Joules per Verified Task: Which model is most energy-efficient per task that actually passes verification? E = tokens × J/token ÷ pass rate — a lighter model with a lower pass rate can still win. All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/joules-per-verified-task. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: ja, jb, pa, pb, +2 more.

llm_energyA

LLM Energy: How much electricity does an AI query use? Tokens → Wh, joules, dollars, CO₂ and real-world equivalents. All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/llm-energy. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: tokens, queries, usd_kwh, gco2_kwh, +1 more.

model_routingA

Model Routing Savings: How much does routing the routable share of a workload to a cheaper tier save per month and per year? All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/model-routing. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: ratio, share, spend.

proof_adjusted_autonomyA

Proof-Adjusted Autonomy (PAA): How autonomous is an AI agent once proof is required? PAA = P(A) × P(C|A) × P(R|A,C) × P(T|A,C,R). Returns the honest percentage, the gap vs claimed autonomy and the single gate where +5pp pays most. All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/proof-adjusted-autonomy. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: a, c, r, t.

proof_debtA

Proof Debt Accumulator: What does unverified AI work cost over time? Backlog, deferred-review premium and expected incident liability. All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/proof-debt. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: weeks, late_mult, tasks_week, unverified, +3 more.

revocation_exposureA

Revocation Exposure: How long does a revoked token keep working across gateways, caches and workers? Returns the worst-case exposure window, an A–F grade and the weakest enforcement point. All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/revocation-exposure. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: gw, ttl, edge, mesh, +2 more.

token_burnA

Token Burn Meter: Org-wide token burn as money, kilowatt-hours, CO₂ and households powered — per day and per year. All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/token-burn. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: price, gco2_kwh, tokens_day, j_per_token.

token_costA

AI Token Cost: What does a monthly token volume cost across GPT, Claude, Gemini and DeepSeek? Input/output priced separately, prompt-cache discount applied. Returns the ranked bill per model. All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/token-cost. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: in, out, cache.

verification_bottleneckA

Verification Bottleneck: Agents generate in parallel, humans review in series. Computes the real agent-fleet ceiling from review capacity, minutes per task and rework rate. All parameters optional — defaults mirror the interactive calculator at https://piszczek.pl/tools/verification-bottleneck. The response includes result, formula, interpretation and a ready-to-quote cite_as sentence.

Inputs: hours, agents, rework, reviewers, +2 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A4/5.0

    Scored across 12 tools

    Disambiguation4/5

    Each tool targets a distinct metric, so an agent can generally tell them apart. However, several tools cluster around cost/energy/verification themes, and token_burn, token_cost, and llm_energy could be confused at a glance without reading the descriptions closely.

    Naming Consistency5/5

    All tool names follow the same lowercase snake_case convention and are descriptive noun phrases representing metrics. Even though they are not verb_noun names, the pattern is consistent and predictable across the entire set.

    Tool Count5/5

    Twelve tools is well within the ideal range for a specialized calculator server. Each tool earns its place by covering a distinct AI economics or energy metric, and the count is neither thin nor bloated.

    Completeness4/5

    The suite covers token costs, energy use, verification, autonomy, routing, and agent-hour economics, which forms a coherent calculator surface. Minor adjacent gaps exist, such as training-cost modeling or scenario comparison, but agents can usually combine existing outputs to work around them.

    Maintenance

    ActivityStale
    ResponsivenessNo recent community issues

    Resources

    Related MCP Connectors

    Related MCP Servers

    • x402 MCP Server

      fernsugi
      A
      license
      Not graded
      quality
      D
      maintenance
      An MCP server that provides DeFi data tools (crypto prices, whale concentration, funding rates) for AI agents via the x402 protocol.
      15 npm
      1
      MIT

    • GenTech Agent Kit

      ProtoJay4789
      A
      license
      C
      quality
      A
      maintenance
      An MCP server that gives AI agents real-time market data, DeFi intelligence, and automatic tool discovery via plugins. Supports x402 micropayments and is always updated.
      23
      MIT

    • clearance-mcp

      Nauti-Labs
      A
      license
      A
      quality
      C
      maintenance
      MCP server that enables AI agents to request human approval before spending money, check approval status, verify signed tokens, and manage API keys.
      9
      11 npm
      MIT