Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    tochi satei kun

    1

    Open-source white-box land valuation engine for Japan using hedonic regression and MLIT transaction data.

    Free

    1 useSecurity scanned
    tochi satei kun

    tochi satei kun

    Example session with this skill installed

    Run a land valuation for a 150sqm lot in Setagaya, 8 mins from the station. I have the MLIT CSV and Setagaya GeoJSON ready in the project folder.

    • Read your context and instructions
    • Compiled the tochi satei kun

    I have processed the land valuation for the specified property in Setagaya. The engine performed a hedonic regression on the provided MLIT transaction data and synchronized the results with the latest Official Land Price benchmarks.

    Valuation Report Summary

    ItemValue
    Subject PropertySetagaya-ku, 150.00 sqm, 8 min to Station
    Valuation Price¥132,000,000 (¥880,000/sqm)
    Price Range¥118,000,000 - ¥145,000,000
    Regression R²0.84 (High Reliability)
    Data Points42 transactions analyzed

    The generated Excel workbook contains the full disclosure of beta coefficients for the 12 features analyzed, including station proximity (β=-0.042) and lot shape (β=0.15). The client-facing sheet has been formatted without technical jargon, suitable for direct presentation.

    
    # Generated File Location:
    # output/土地査定_世田谷区_20240520.xlsx
    

    Next steps

    1. Review the Pro Sheet to verify the hedonic coefficients and p-values for local market fit.
    2. Check the

    Comparison Table in the workbook to see the top 3 similar transactions used for individual factor adjustments.
    3. Present the Client Sheet to the seller to justify the suggested listing price range.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Pricing land in Japan is often a "black box" of hidden adjustments. This skill provides an open-source, white-box Automated Valuation Model (AVM) designed for real estate professionals. It transforms raw MLIT transaction data and Official Land Price GeoJSON files into professional, explainable appraisal reports.

    What it does

    • Statistical valuation calculates land prices using hedonic regression across 12 specific features including area, shape, and station proximity.
    • Explainable logic discloses all beta coefficients, p-values, and standard errors, ensuring the valuation isn't just a number but a defensible calculation.
    • Dual-sheet output generates a detailed "Pro" sheet for internal review and a simplified "Client" sheet compliant with industry standards.
    • Comparable selection identifies top-weighted similar transactions and aligns them using weighted distance scores.
    • Automated adjustment applies time corrections based on official land price trends and individual factor corrections for specific lot characteristics.

    How it works

    1. The user provides a property profile and two required data files (MLIT CSV and Official Land Price GeoJSON).
    2. The skill executes scripts/main.py to run the full valuation pipeline, from data scoping to regression analysis.
    3. The system generates a three-sheet Excel workbook (.xlsx) containing the Pro, Client, and Supplemental Data sheets.

    Frameworks & tools

    Built with Python, utilizing statsmodels for OLS regression, openpyxl for Excel generation, and specialized scripts for geospatial data handling and MLIT parsing.

    Why this beats prompting it yourself

    Generic prompts cannot handle regression analysis of thousands of transaction records or calculate precise time-adjustments from GeoJSON files. This skill uses a hard-coded mathematical pipeline that ensures every yen of valuation is backed by a specific coefficient rather than an LLM guess.

    Use cases

    • Generating initial valuation reports for prospective sellers during the listing phase.
    • Providing objective evidence for price negotiations based on historical transaction data.
    • Comparing subject property values against official land price (Koji Chika) benchmarks.
    • Creating white-labeled valuation documents for real estate agency clients.

    Known limitations

    Supports only vacant land and ownership rights in Japan. Buildings, apartments, and leasehold rights are excluded. Requires external CSV/GeoJSON data files as inputs.

    Supports only Japanese vacant land (ownership). Excludes buildings, apartments, and leaseholds. Requires manual download of MLIT CSV and GeoJSON files. Minimum 15 data points for regression.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    Unlike generic AI models that guess prices, this skill runs a rigorous OLS regression on thousands of MLIT records, providing you with the exact coefficients and p-values used to reach the valuation for full professional transparency.

    The skill is designed for agents capable of executing Python environments and handling file uploads; it requires two specific datasets—MLIT transaction CSVs and Official Land Price GeoJSON files—to perform its analysis.

    This skill is specialized for vacant land and ownership rights (shoyuken) within Japan; it does not currently support building valuations, apartments, or leasehold (shakuyuken) interests.

    You will receive the complete Python-based valuation pipeline, including the regression scripts, data parsing tools, and the Excel generation templates, all ready to be integrated into your AI workflow.

    Yes, the skill generates a dual-sheet Excel workbook featuring a 'Client' sheet formatted for professional presentations and a 'Pro' sheet containing the technical statistical data for internal verification.

    New in Trading & Investing