Read a filing's text
get_filing_text is an MCP tool that returns the narrative sections of an annual securities report (有価証券報告書) exactly as the company filed them. The text is Japanese, reproduced verbatim from the filing’s text blocks with markup removed. No AI writes or rewrites it.
What you get
Section titled “What you get”Two sections are available:
business_risks— 事業等のリスク, the risks the company reports about its business.md_and_a— the management analysis of financial position, operating results and cash flows.
If you do not name a section, you get business_risks. Each returned section carries its Japanese and English label, the XBRL element it was read from and its length. Long text comes in pages; ask Claude or ChatGPT to continue to read the next page. A section that the filing does not hold is listed as missing.
How to ask
Section titled “How to ask”Connect Kabuken to your AI assistant first. See Connect the EDINET MCP server to Claude and ChatGPT. Then ask in plain language, in English or Japanese:
Show me the business risks section of Sony Group’s latest annual report, in the original Japanese.
ニレコの有価証券報告書の事業等のリスクを原文で見せてください。
Show the management analysis (経営者による財政状態、経営成績及びキャッシュ・フローの状況の分析) from Toyota’s latest annual securities report.
Which filing is used
Section titled “Which filing is used”Give a company and the tool uses its newest annual report that has XBRL data. To read a specific filing, give its EDINET document id (for example one returned by list_documents) and ask for that filing. The result names the filing it was read from by its EDINET document id (doc_id) and its period end (period_end), so you can check it on EDINET.
Rights
Section titled “Rights”Text is reproduced verbatim from the filing; the filing company holds the rights to its text. Takedown requests: https://kabuken.intellics.ai/support