{
  "markdown": "# ai-economics-mcp\n\n[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.21760145.svg)](https://doi.org/10.5281/zenodo.21760145)\n\n[![npm](https://img.shields.io/npm/v/%40michalpiszczek%2Fai-economics-mcp?label=npm)](https://www.npmjs.com/package/@michalpiszczek/ai-economics-mcp) [![Glama score](https://glama.ai/mcp/servers/pich/ai-economics-mcp/badges/score.svg)](https://glama.ai/mcp/servers/pich/ai-economics-mcp) [![MIT](https://img.shields.io/badge/license-MIT-green)](LICENSE)\n\n**MCP server for AI economics.** Gives Claude, Cursor and any MCP client 12 calculators for\nthe questions that decide AI budgets: what tokens cost, what queries burn, and whether agent\nwork can be proven.\n\nListed in the official [MCP Registry](https://registry.modelcontextprotocol.io) as **`pl.piszczek/ai-economics`** (domain-verified). Wraps the free [piszczek.pl/tools](https://piszczek.pl/tools) API — no key, no sign-up,\nstateless (inputs are never stored). The math is the same as the interactive calculators.\n\n## Install\n\n**Claude Desktop / Claude Code** (`claude_desktop_config.json` / `.mcp.json`):\n\n```json\n{\n  \"mcpServers\": {\n    \"ai-economics\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@michalpiszczek/ai-economics-mcp\"]\n    }\n  }\n}\n```\n\n**Claude Code one-liner:**\n\n```bash\nclaude mcp add ai-economics -- npx -y @michalpiszczek/ai-economics-mcp\n```\n\n## Tools\n\n| Tool | Answers |\n|---|---|\n| `token_cost` | What do 1M tokens cost per month across GPT, Claude, Gemini, DeepSeek? |\n| `context_window` | How many tokens is 50 pages, does it fit, what does carrying it cost? |\n| `agent_hour` | What does one hour of an AI agent cost, fully loaded with verification? |\n| `model_routing` | How much does routing to a cheaper tier save per year? |\n| `llm_energy` | How much electricity does an AI query use? |\n| `joules_per_verified_task` | Which model is most efficient per task that actually passes? |\n| `token_burn` | What does org-wide token burn cost in money, kWh, CO₂? |\n| `humanoid_energy` | How long can a humanoid robot run per charge? |\n| `proof_adjusted_autonomy` | How autonomous is a \"90% autonomous\" agent once proof is required? |\n| `revocation_exposure` | How long does a revoked token keep working across your stack? |\n| `verification_bottleneck` | How many AI agents can a human review team absorb? |\n| `proof_debt` | What does unverified AI work cost over time? |\n\nAll parameters are optional; defaults mirror the interactive calculators. Every response\nincludes `result`, `formula`, `interpretation` and a ready-to-quote `cite_as` sentence.\n\n## Example\n\n> **You:** how autonomous is our agent really? it completes 90% alone, evidence for 95%,\n> 80% independently validated, 90% on time\n>\n> **Claude** (calls `proof_adjusted_autonomy`): PAA = 61.6% — supervised autonomy. The gap\n> vs the claimed 90% is 28.4 pp, and the biggest lever is independent validation…\n\n## Configuration\n\n| Env var | Default | Purpose |\n|---|---|---|\n| `AI_ECONOMICS_API` | `https://piszczek.pl/tools/api` | Point at a self-hosted instance |\n\n## Glama builds and releases\n\nGlama generates its own Dockerfile from the\n[build configuration](https://glama.ai/mcp/servers/pich/ai-economics-mcp/admin/dockerfile).\nUse these settings:\n\n- Build steps: `[\"npm ci --omit=dev\"]`\n- CMD arguments: `[\"mcp-proxy\", \"--\", \"node\", \"index.js\"]`\n- Environment variables: optional `AI_ECONOMICS_API`; no credentials required.\n\nThis is plain JavaScript: there is no `build` script or compilation step. Do not use\n`pnpm run build` or `npm run build`. After syncing the repository, build the selected\ncommit and publish a Glama release from the successful test. A GitHub or npm release\nalone does not trigger Glama's quality evaluation.\n\nFor local stdio use, the repository Dockerfile needs no proxy or exposed ports:\n\n```bash\ndocker build -t ai-economics-mcp .\ndocker run -i --rm ai-economics-mcp\n```\n\n## Concepts behind the tools\n\n- [Joule Wars](https://piszczek.pl/joule-wars) — the AI race for energy efficiency\n- [Proof-Adjusted Autonomy](https://piszczek.pl/proof-adjusted-autonomy) — the metric of proven agent work\n- [Revocation Exposure](https://piszczek.pl/glossary/revocation-exposure) — how long revoked authority keeps working\n\n## Citing a result\n\nEvery 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](https://piszczek.pl/cite).\n\n## License\n\nMIT. Concepts and calculators by [Michał Piszczek](https://piszczek.pl/michal-piszczek)\n(CC BY 4.0 — attribution appreciated).\n",
  "bytes": 4635,
  "sha": "decea750f17a34f616c7756dbb72fb9027991a5589439a6509d17f433e30563f",
  "repo_slug": "pich/ai-economics-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_pl_piszczek_ai_economics_ad1b3f20/readme"
}