# AutonoMath (English index)

> Cross-search REST + MCP API for Japanese public programs: 10,790 active subsidies / loans / tax measures / certifications (tier S/A/B/C searchable; tier X = 1,923 quarantine excluded) + 2,286 adoption case studies + 108 loan products (collateral / personal-guarantor / third-party-guarantor decomposition) + 1,185 enforcement cases + 9,484 laws + 35 tax rulesets + 2,065 court decisions + 362 bids + 13,801 invoice registrants + 503,930 entities / 6.12M facts / 23,805 relations / 335,605 aliases, 181 exclusion/prerequisite rules, primary-source URL + fetched_at on 99%+ of rows. MCP-native (72 tools, protocol 2025-06-18). ¥3 per request metered (¥3.30 tax-inclusive), anonymous 50 req/month per IP free (JST first-of-month reset).

This file follows the emerging llms.txt convention (https://llmstxt.org/) —
a curated, plaintext index of canonical documents for LLM indexers.
The Japanese-language sibling lives at https://autonomath.ai/llms.txt.
A flat-root mirror of this English index is at https://autonomath.ai/llms.en.txt.
Concatenated full docs: https://autonomath.ai/llms-full.en.txt
Canonical domain: https://autonomath.ai (Bookyou Inc. / corporate number T8010001213708 / info@bookyou.net).

Program names, law titles, and case-study fields are kept verbatim in
Japanese. They are legal entity names; machine-translating them would
invent canonical names that do not exist (a fraud-risk vector under
Japanese 景表法 / 消費者契約法). The structural prose around the data is
in English so English-language LLM crawlers can index the framing.

## Operator

Bookyou Inc. (Japanese corporate number T8010001213708) — Representative: Shigetoshi Umeda — info@bookyou.net

## Docs

- [Overview](https://autonomath.ai/docs/)
- [Getting started (5-minute quickstart)](https://autonomath.ai/docs/getting-started/): install, API key, curl/Python/Node, Claude Desktop MCP config
- [API reference (REST)](https://autonomath.ai/docs/api-reference/)
- [MCP tools (72)](https://autonomath.ai/docs/mcp-tools/): 39 jpintel + 33 autonomath stdio tools
- [Examples (Python + curl)](https://autonomath.ai/docs/examples/)
- [Error handling](https://autonomath.ai/docs/error_handling/)
- [Exclusion rules](https://autonomath.ai/docs/exclusions/)
- [FAQ](https://autonomath.ai/docs/faq/)
- [Pricing (¥3 per request metered)](https://autonomath.ai/docs/pricing/)
- [SLA](https://autonomath.ai/docs/sla/)

## Examples

- [Examples (Python + curl)](https://autonomath.ai/docs/examples/)
- [Try the prescreen widget](https://autonomath.ai/en/index.html#try)

## MCP server

- PyPI package: `autonomath-mcp` — `pip install autonomath-mcp`
- Console script: `autonomath-mcp` (FastMCP stdio, 72 tools at default gates)
- Repo: https://github.com/Bookyou/autonomath-mcp

## Data licenses

- e-Gov law corpus: CC-BY 4.0
- NTA invoice-registrant bulk: PDL v1.0 (Public Domain License, redistribution OK with attribution)
- gBizINFO corporate metadata: Japan Government Standard Terms of Use
- Public-domain rows (gazette etc): public domain
- Each row carries an explicit license via `am_source.license` (99.997% coverage)
- Full terms: https://autonomath.ai/en/tos.html

## Structured data

- [Sitemap index](https://autonomath.ai/sitemap-index.xml)
- [Docs sitemap](https://autonomath.ai/docs/sitemap.xml)
- [Programs sitemap](https://autonomath.ai/sitemap-programs.xml)
- Per-program JSON-LD (schema.org GovernmentService / LoanOrCredit) is embedded inline on each `/programs/{id}.html`

## Optional

- [About](https://autonomath.ai/en/about.html)
- [Pricing page (HTML)](https://autonomath.ai/en/pricing.html)
- [Terms of service](https://autonomath.ai/en/tos.html)
- [Privacy policy](https://autonomath.ai/en/privacy.html)
- [Security disclosure (RFC 9116)](https://autonomath.ai/.well-known/security.txt)