Japan Company Disclosure Monitor — 適時開示・法人情報 API

Monitor Japanese company disclosures and query company intelligence in one call: TDnet timely disclosures (適時開示), EDINET filings (有価証券報告書・大量保有報告書), and gBizINFO corporate registry data — unified into clean JSON.

Built for scheduled watchlists, analysts, investors, n8n/Dify workflows, and AI agents that need structured Japanese market data without navigating Japanese-only government portals.

What you get

Source	Data	Key required?
TDnet (Tokyo Stock Exchange)	Timely disclosures: earnings revisions, buybacks, M&A, executive changes, dividend announcements	No
EDINET (Financial Services Agency)	Official filings: annual/quarterly securities reports, large shareholding reports	Free key (bring your own)
gBizINFO (METI)	Corporate registry: corporate number, address, representative, capital, employees	Free token (bring your own)

All sources are official government/exchange data — no scraping of private websites, no ToS gray zones.

Modes

1. Watchlist monitor (recommended)

Monitor selected TSE securities codes over a recent time window. Use multiple title keywords with OR matching, or leave keywords empty to return every disclosure for the watchlist.

{
    "mode": "watchlist",
    "watchlist": ["7203", "6758", "8306"],
    "keywords": ["業績予想", "自己株式", "M&A"],
    "sinceHours": 24,
    "maxDisclosures": 50
}

Each TDnet record includes a stable disclosureId, so schedulers, webhooks, n8n, Dify, and AI agents can deduplicate repeated lookback windows safely.

2. Company report

Query one company by name (トヨタ自動車) or 4-digit securities code (7203) and get its recent disclosures, filings, and registry profile.

{
    "mode": "company_report",
    "query": "7203",
    "maxDisclosures": 30
}

3. Market feed

Get the latest disclosures across the whole market, optionally filtered by one keyword — e.g. every 自己株式 (buyback), 上方修正 (upward revision), or MBO announcement.

{
    "mode": "market_feed",
    "keyword": "自己株式",
    "maxDisclosures": 50
}

Schedule a disclosure monitor

1. Run the Actor once with mode: watchlist and confirm the records you want.
2. Save the input as an Apify Task.
3. Add an hourly or daily Schedule. For hourly schedules, use sinceHours: 2 to provide overlap.
4. Connect the finished run or dataset to an Apify webhook/integration, then deduplicate downstream on disclosureId.

The Actor intentionally returns overlapping records rather than keeping shared hidden state. This makes scheduled runs reproducible and prevents one user's watchlist state from affecting another user's results.

Output example

{
    "type": "tdnet_disclosure",
    "disclosureId": "tdnet_2f4f8d07d9f8a2b73c10",
    "source": "TDnet (official site)",
    "companyCode": "72030",
    "companyName": "トヨタ自動車",
    "title": "自己株式取得に係る事項の決定に関するお知らせ",
    "pubdate": "2026-07-27 15:00",
    "publishedAt": "2026-07-27T06:00:00.000Z",
    "watchlistMatched": true,
    "matchedKeywords": ["自己株式"],
    "documentUrl": "https://www.release.tdnet.info/...pdf"
}

A report_summary item is always appended with source counts and any warnings.

Pay-per-event note: Apify charges each default dataset item as one result event. Therefore a run with 5 disclosure records produces 6 billable result events: 5 disclosures plus 1 report_summary. A zero-result run still produces the summary so that automation receives a structured status.

Why bring your own EDINET/gBizINFO keys?

Both keys are free and take ~5 minutes to obtain:

• EDINET: https://api.edinet-fsa.go.jp/ (email registration)
• gBizINFO: https://info.gbiz.go.jp/hojin/api_registration/form

The Actor works without them (TDnet-only mode), but adding them unlocks the full report. Keys are marked secret and never stored.

Notes & limitations

• Company-name queries against TDnet match recent disclosures only; securities codes are more reliable.
• gBizINFO lookup requires a company name (the registry has no securities-code index).
• English company names are not yet supported for TDnet queries — use the securities code instead.
• This Actor returns public regulatory data as-is. It is not investment advice.

Use cases

• AI agents / RAG: feed structured Japanese disclosure data into LLM pipelines (MCP-friendly output).
• Investor monitoring: schedule daily runs with keyword filters (buybacks, revisions, MBO) and pipe to Slack/email via Apify integrations.
• Due diligence: one-call company snapshot combining exchange, regulator, and registry data.
• Market research: track disclosure activity across the Japanese market without reading Japanese portals.

Support

Open an issue on the Actor's Issues tab — maintained actively, typical response within 1–2 business days.