langchain-axiora 0.2.1

LangChain tools for Japanese financial data — EDINET filings, XBRL, company analysis (Axiora API)

langchain-axiora

LangChain tools for Japanese financial data — EDINET filings, XBRL, and company analysis.

Give your LangChain / LangGraph agents access to 4,000+ Japanese listed companies: financials from EDINET filings, health scores, screening, and English translations of annual reports. Powered by the Axiora API.

Install

uv add langchain-axiora
# or
pip install langchain-axiora

Quick Start

from langchain_axiora import AxioraToolkit
from langchain_anthropic import ChatAnthropic
from langgraph.prebuilt import create_react_agent

toolkit = AxioraToolkit(api_key="ax_live_...")
agent = create_react_agent(ChatAnthropic(model="claude-sonnet-4-20250514"), toolkit.get_tools())

result = agent.invoke({
    "messages": [{"role": "user", "content": "Compare Toyota and Honda financials"}]
})

Configuration

The API key can be provided three ways (in priority order):

# 1. Explicit
toolkit = AxioraToolkit(api_key="ax_live_...")

# 2. Environment variable
import os
os.environ["AXIORA_API_KEY"] = "ax_live_..."
toolkit = AxioraToolkit()

# 3. .env file (with python-dotenv)
from dotenv import load_dotenv
load_dotenv()
toolkit = AxioraToolkit()

Using a Subset of Tools

18 tools can be noisy for simple agents. Pass selected_tools to limit:

toolkit = AxioraToolkit(
    api_key="ax_live_...",
    selected_tools=[
        "axiora_search_companies",
        "axiora_get_financials",
        "axiora_compare_companies",
    ],
)
# Invalid names raise ValueError with the valid list

Using Individual Tools

Tools accept api_key directly — no wrapper needed:

from langchain_axiora import GetFinancialsTool

tool = GetFinancialsTool(api_key="ax_live_...")
result = tool.invoke({"code": "7203", "years": 3})

# Or use AXIORA_API_KEY env var
tool = GetFinancialsTool()

# Async
result = await tool.ainvoke({"code": "7203"})

Available Tools (18)

Tool	Description
axiora_search_companies	Search companies by name or code
axiora_get_company	Get detailed company info
axiora_get_financials	Revenue, income, assets, ROE over time
axiora_get_growth	YoY growth rates and CAGRs
axiora_get_ranking	Rank companies by any metric
axiora_get_sector_overview	Sector stats and company counts
axiora_compare_companies	Side-by-side comparison (2-5 companies)
axiora_screen_companies	Filter by revenue, ROE, PE ratio, sector
axiora_get_health_score	Financial health score (0-100)
axiora_get_health_ranking	Rank by health score
axiora_get_peers	Find similar companies in same sector
axiora_get_timeseries	Chart-ready metric time series
axiora_list_filings	List annual/quarterly filings
axiora_get_translations	English translations of filings
axiora_search_translations	Full-text search across translations
axiora_get_filing_calendar	Filing activity per day
axiora_search_companies_batch	Bulk company lookup
axiora_get_coverage	Data coverage statistics

Retriever

Search English translations of Japanese filings as LangChain Documents:

from langchain_axiora import AxioraRetriever

retriever = AxioraRetriever(api_key="ax_live_...", section="risk_factors", k=5)
docs = retriever.invoke("semiconductor supply chain risk")

for doc in docs:
    print(doc.metadata["company_name"], doc.page_content[:100])

Works in any LangChain chain or RAG pipeline that accepts a retriever.

Error Handling

All tools use handle_tool_error=True. When the API returns an error, the agent receives a helpful message instead of crashing:

Axiora API error 404. Company not found. Use axiora_search_companies to find the correct code.

This lets the agent self-correct (e.g., search for the right code, then retry).

Get an API Key

Sign up at axiora.dev — free, no credit card.

Docs · PyPI