Search Japanese subsidies & tax rules with one line of AI.

¥3.30/req (tax incl.) · 50 req/month free, no signup. Drop into Claude / Cursor / curl and run.

Try free (50 req/month per IP, no signup) Add to Claude Desktop (.mcpb)

Anonymous 50 req/month per IP free (JST UTC+9 first-of-month reset); ¥3/req tax exclusive (¥3.30 tax inclusive) metered beyond. Primary-source URL + fetched_at on 99%+ of rows.

Data & quality

503,930 entities + 6.12M facts

9 datasets + entity-fact DB, one query surface

Programs, adoption case studies, loans, enforcement, laws, court decisions, tax rulesets, bids, invoice registrants — all under a unified schema. The entity-fact layer carries entities and facts across 12 record kinds. The full-text search index (3-gram) lets you query Japanese as written, no romanization required.

  • programs · 14,472 (11,684 searchable = Tier S 114 + A 1,340 + B 4,186 + C 6,044; 2,788 held back for secondary review)
  • case_studies · 2,286
  • loan_programs · 108 (collateral / personal-guarantor / third-party-guarantor decomposed)
  • enforcement_cases · 1,185
  • laws · 154 full-text indexed + 9,484 catalog stubs (e-Gov, CC-BY; body-text load incremental, name resolver covers all 9,484)
  • tax_rulesets · 35
  • court_decisions · 2,065
  • bids · 362
  • invoice_registrants · 13,801 (NTA PDL v1.0 delta; full monthly bulk pending)
  • entity-fact DB · 503,930 entities / 6.12M facts / 23,805 relations / 335,605 aliases

99%+ primary source

source_url + fetched_at on 99%+ of rows

Sourced directly from METI, MAFF, the SME Agency, Japan Finance Corporation, prefectures, and municipalities. Aggregator sites are excluded by policy. Designed so amendments propagate through immediately.

  • noukaweb / hojyokin-portal etc. excluded
  • Nightly source-URL liveness scan
  • Lineage column required on every row

181 exclusion rules

Tier S/A/B/C/X quality labels

Co-use conflicts, prerequisites, and mutual exclusivity are encoded as structured rules. Post a list of program_ids, get back the conflicts. No "plausible-sounding" LLM inference — actual rules.

  • 125 exclude + 17 prerequisite
  • 15 absolute + 24 other
  • Tier S/A/B/C = data-completeness ranks; X = quarantine

Five interfaces

One backend, five ways to deliver it. Pick what fits the audience.

API + MCP

AI developers / enterprise RAG

Call 89 MCP tools directly from an agent. REST is OpenAPI 3.1. stdio transport, no SDK.

  • ¥3/req tax exclusive (¥3.30 tax inclusive)
  • Anonymous 50 req/month per IP free
  • Claude Desktop / Cursor / ChatGPT compatible

Read the docs →

LINE bot

SMB owners / professional advisors

Discover subsidies in 30 seconds, in plain Japanese, on the LINE app you already have. Add as a friend, no app install.

  • ¥3/notification tax exclusive (¥550 tax inclusive)
  • Up to 10 questions/month free
  • Apple Pay / Google Pay supported

Use the LINE bot →

Law-amendment alerts

Tax / labor / compliance professionals

Email notifications on amendments to invoice law, electronic-bookkeeping law, labor standards, and other tracked domains. Catch changes before deadlines bite.

  • ¥3/notification tax exclusive (¥550 tax inclusive)
  • Monthly digest is free
  • Real-time notifications are paid

Configure alerts →

Embeddable widget

Tax-firm sites / chambers of commerce / advisory offices

Add subsidy search to your own site with a single <script> tag. Visitors narrow down candidates without leaving your page.

  • ¥3/search metered (10,000 req included)
  • Overage ¥3/req
  • White-label ¥3/search metered

Embed the widget →

Advisor affiliate

Certified support orgs / tax / labor / SMB consultants

Marketplace that routes inbound inquiries to certified support institutions and licensed advisors. Advisors pay only on successful engagements.

  • Free for end users
  • Advisor pays ¥3,000 per signed engagement
  • No monthly fixed fee

Register as an advisor →

Not sure which fits?

Comparison table + decision tree

Individual / corporate / developer / advisor — a 5-question decision tree picks the right interface for you. Typical use cases and setup time line up side by side.

  • Five-interface comparison table
  • Decision tree per audience
  • Combo recommendations (e.g. MCP + LINE)

See the comparison →

Five ways to use it — pitch by audience

Same backend, different framing for each role.

Tax advisors (zeirishi)

Mid-size firm, ~80 SMB clients

When a client asks "which subsidies apply to us this year?", point Claude at the API. About ¥3,000/month metered. Email digest of monthly amendments included.

  • API + MCP (¥3/req)
  • Law-amendment alerts (¥3/notification)
  • Monthly digest is free

Tax advisor details →

Administrative scriveners (gyoseishoshi)

Construction-focused, permit-heavy caseload

Pull "applicable subsidies + loans + permits" for a case in one call. Filter by region × industry × deadline. Try the first 50/month free.

  • API + MCP — one call, cross-source
  • Anonymous 50 req/month per IP free
  • Region × industry × deadline search

Admin scrivener details →

SMB owners

Small business — owner, spouse, one bookkeeper

Just ask on LINE: "are there subsidies for my industry?" Up to 10 questions/month free, ¥3 per question after that. Zero setup.

  • LINE bot — add as friend, that's it
  • Up to 10 questions/month free
  • ¥3 per question (tax exclusive) after

SMB details → · LINE bot

VC / M&A advisors

Investors / acquirers running a DD pipeline

From a corporate number, fetch 5 years of enforcement history, 10 years of adoption history, and invoice-registrant status in one query. Drop straight into your DD pipeline. ¥3/req metered, no scale ceiling.

  • One query keyed on corporate number
  • Enforcement 1,185 / adoptions 2,286 / invoice 13,801
  • ¥3/req fully metered, no scale limits

VC / DD details →

AI agent developers

Claude Desktop / Cursor / your own SaaS integration

A Japanese subsidy / tax / loan MCP server. 89 tools. ¥3/req, 50/month free. Works in Claude Desktop and Cursor out of the box.

  • 89 MCP tools, protocol 2025-06-18
  • stdio transport, no SDK
  • 50 req/month evaluation, then ¥3/req metered

Developer details → · MCP tool list

Not on this list?

If none of the five fit

Labor consultants, SMB consultants, accounting SaaS, chambers of commerce, municipal officers — the decision tree finds them too. Compare typical use cases, combo recommendations, and setup time side by side.

  • 5-question decision tree
  • Combo recommendations (MCP + LINE etc.)
  • Typical use cases lined up

Audience comparison →

Try in 5 seconds (no signup)

Anonymous IP gets 50 req/month free. Paste and run.

$ curl 'https://api.jpcite.com/v1/programs/search?q=IT%E5%B0%8E%E5%85%A5&limit=3'
{
 "results": [...],
 "total": 17
}
Query is URL-encoded (q=IT導入).

Using an MCP client? → Claude Desktop integration guide · Cursor · Cline

This is a search index. We do not provide legal advice, tax advice, or filing services. Activities reserved under 弁護士法 § 72 / 税理士法 § 52 / 行政書士法 § 1 are out of scope. See error_handling · pricing · Terms.

Subsidy Prescreen — Top 5 in 3 questions

Pick prefecture / business form / planned investment. We'll show the top 5 of 11,684 active programs ranked by fit. Free, no signup, anonymous (50 reqs/month/IP). One call to /v1/programs/prescreen.

Picking "Nationwide" drops the regional filter and searches every program. Picking a specific prefecture matches its prefecture-only programs plus all nationwide programs.
"Unspecified" matches only programs accepting both forms. Picking "Corporation" or "Sole proprietor" returns form-specific programs plus shared ones.
Planned investment for equipment, IT systems etc., in 万円 units (optional). Example: 800 = 8,000,000 yen. Programs with ceilings near your amount are ranked higher.

Matching is rule-based against the program database, not LLM inference (prefecture + target_types + amount_fit). Source URLs are primary only. Submissions are not stored — only an IP hash is logged for rate limiting (resets at JST first-of-month).

Why not just ask ChatGPT?

Short answer: no. Three structural reasons LLMs alone can't deliver fact-grade Japanese-program judgments.

Stale training data

Calendar mismatch

ChatGPT/Claude/Gemini training cutoffs run months-to-a-year stale. Japanese subsidies move on calendar deadlines ("application closed yesterday"); an LLM alone can't tell you what's currently open. jpcite ships fetched_at on every row.

Source fabrication & aggregator drift

Misidentification = misrepresentation liability

Aggregator sites (noukaweb, hojyokin-portal, biz.stayway, etc.) are documented to list expired programs as active, mis-state ceiling amounts, and add eligibility requirements that aren't in the primary notice. Filing an application based on aggregator data and being rejected exposes the filer to misrepresentation liability from their own client.

jpcite bans aggregator URLs from source_url as a hard policy. All sources are primary: ministries (METI / MAFF / SME Agency), prefectures, government finance corps (JFC / SMRJ / NEDO), and chambers of commerce. 99%+ of 14,472 rows carry a primary URL; the remaining 12 (small-municipality programs) are flagged "primary URL missing" rather than backfilled with aggregators.

Every row carries source_url + fetched_at, so you can show your client / regulator / opposing counsel exactly why this program matched. That audit trail isn't constructible if aggregators are in the chain.

No compatibility logic

Combination rules can't be inferred

LLMs can't structurally answer "can I combine subsidy A with subsidy B?" or "does this conflict with my prior award?". jpcite ships 181 exclusion / prerequisite rules for machine-decidable compatibility.

In short: LLM agent + jpcite = LLM speed × primary-source accuracy. jpcite isn't an LLM replacement — it's the fact layer that makes LLM output trustworthy.

Why jpcite

No aggregator sources

Misrepresentation-liability avoidance

Aggregator sites such as noukaweb, hojyokin-portal, and biz.stayway have been documented to keep expired programs listed as active, mis-state ceiling amounts, and add eligibility requirements that aren't in the primary notice. Filing on that basis and being rejected exposes the filer to misrepresentation liability from their own client.

jpcite's policy is to never register an aggregator URL in source_url. Sources are restricted to ministries (METI / MAFF / SME Agency), prefectures, government finance corps (JFC / SMRJ / NEDO), and chambers of commerce. 99%+ of 14,472 rows carry a primary URL; the remaining 12 (small-municipality programs without a dedicated CMS page) are flagged "primary URL missing" rather than backfilled.

Because every row carries source_url + fetched_at, you can show your client, the regulator, or opposing counsel exactly why a program matched. That audit trail isn't constructible if aggregators are in the chain.

Primary source, end to end

Amendments propagate immediately

Every program row carries source_url + fetched_at. The fetch date is shown verbatim — never relabeled as "last updated" (which would falsely imply we re-verified currency). Nightly liveness scans flag dead URLs.

Exclusion logic is in the data, not the prompt

Catch ineligible combinations up front

181 exclusion / prerequisite / absolute rules, structured. The "agent merrily files two incompatible subsidies, both get rejected months later" failure mode is detected the moment you pass program_ids in.

Cross-dataset glue

One call: program → law → court decision → enforcement

Expansion datasets are live: laws (e-Gov, CC-BY; 154 rows full-text indexed + 9,484 catalog stubs, body-text load incremental, name resolver covers all 9,484) + tax rulesets (invoice law / electronic-bookkeeping law; 35 rows) + NTA invoice registrants (PDL v1.0; 13,801 rows live delta) + court decisions (2,065 rows) + bids (362 rows). Plus the autonomath entity-fact DB (503,930 entities / 6.12M facts / 23,805 relations) exposed via 30 tools. The 69-tool MCP surface includes trace_program_to_law / combined_compliance_check / search_tax_incentives / reason_answer for cross-domain resolution.

Who is it for

Developers / AI engineers

Recommended: API + MCP

For embedding into agents, RAG pipelines, or your own SaaS. OpenAPI 3.1, stdio MCP, ¥3/req metered, anonymous 50 req/month for evaluation with no signup.

SMB owners

Recommended: LINE bot + advisor affiliate

Want to know what your business is eligible for, and to be routed to someone who can file it. Discover via LINE, get matched with a certified support institution or licensed tax advisor. The first 10 questions/month are free; ¥3/question (¥3.30 tax incl.) thereafter — no fixed monthly fee.

Professional advisors (tax / labor / SMB consultants)

Recommended: alerts + embeddable widget + advisor registration

Don't miss amendments, generate inbound on your own site, accept routed cases. Alerts ¥3/notification, widget ¥3/search metered, advisor billing only on signed engagements.

Corporate (compliance / accounting)

Recommended: alerts + API for internal RAG

Stay current on invoice law, electronic-bookkeeping law, and Premiums & Misrepresentation Act amendments. Alerts surface the change; the API serves the detail to internal tooling.

Pricing summary

Interface Free tier Paid Tax inclusive
API + MCP 50 req/month per IP ¥3 / req (tax excl.) ¥3.30 / req
LINE bot 10 questions/month ¥3 / question (tax excl.) ¥3.30 / question
Law-amendment alerts Monthly digest ¥500 / month (tax excl.) ¥550 / month
Embeddable widget — ¥10,000 / month (10,000 req included) ¥11,000 / month
Advisor affiliate Free for end users ¥3,000 / signed engagement (paid by professional) ¥3,300 / engagement

Full pricing details → · Auto-issued tax invoice , Stripe Tax handled, cancel anytime. Billing is JPY; 10% JCT added at invoice; your card network handles FX.