Connect Your AI

Your connect URL

Add this URL to your assistant as a custom connector. It gives the assistant read-only lookups over the same numbers this site publishes: ~820 Japanese CPI series, monthly from January 1970 to the latest published month.

Set up in your client

claude.ai (web)

1. Open Settings → Connectors.
2. Choose Add custom connector.
3. Name it Japan Data Observatory, paste the URL above, and add it.
4. In a new chat, enable the connector, then ask a question from the list below.

Claude Desktop

1. Open Settings → Connectors.
2. Choose Add custom connector, paste the URL above, and add it.

Claude Code

One command in a terminal:

Other clients

Any client that supports remote MCP servers over Streamable HTTP can use the same URL. The server is stateless and requires no authentication.

Once connected, the full manual covers every lookup, how to read the trust labels in what comes back, and what to do when something looks wrong.

What to ask

• What is Japan's headline and core inflation right now?
• Which expenditure groups are driving headline inflation, in percentage points?
• How has the price of rice moved over the last three years?
• How broad is inflation across individual items — what share is rising more than 2%?
• Compare electricity and city gas prices since 2022.

What your assistant can see

Read-only lookups, each a thin wrapper over this site's public API — the assistant can only see numbers the site already publishes, from the same accepted release. Six generic tools cover every dataset on the server:

• List datasets — every dataset: what it is, its shape, and which tools apply.
• Describe a dataset — its source, every measure with its trust label and formula, and what this server holds.
• Search — a company by name or securities code, or a series by name or code, across all datasets.
• Get a company — one dataset's full view, or a profile across every dataset that knows the company.
• Get series history — published values or calculated rates, as of any date.
• Ranked screen — the dataset's own screens: oldest boards, largest unspent buybacks, cheapest land, lowest director approval, highest ROE.

The dataset cards are also readable as MCP resources. The older per-dataset tools remain available for compatibility. Consumer prices:

• List datasets — the published tables and what each covers.
• Search series — find any of the ~820 series by English or Japanese name.
• Get series history — index levels or rates of change for up to six series.
• Inflation overview — headline, cores, and the ten major groups at the latest month.
• Contributions to headline — the percentage-point decomposition by group.
• Inflation breadth — the share of individually priced items rising or falling.

Cross-shareholdings, where that dataset is published:

• Cross-shareholding coverage — filers extracted, named holdings, total book value.
• Search companies — by Japanese name or securities code.
• Company holdings — what a company holds and who holds it, exactly as filed.
• Unwind ranking — filers by policy-holding value, with positions cut versus added.

The manual documents each one in detail.

Trust and limits

• Index levels are official statistics exactly as the Statistics Bureau of Japan published them; year-over-year and other rates are calculated by this platform, and every lookup result carries that label plus the formula used.
• Every result includes a permanent citation URL — the page on this site showing the same view — so any figure the assistant quotes can be checked at the source.
• Missing values are returned as missing, never as zero.
• The assistant's prose is the product of the AI client you connect, not of this site. Check its statements against the cited pages before quoting them; the Methodology page explains every label and formula.
• The endpoint is rate-limited per connection. For bulk or programmatic access, use the public API directly.