使用方法
- 复制安装命令,在自己的智能体环境安装前先检查源码包。
- 从原始文档核对所需工具、凭据和支持的宿主。
- 说明输入材料与期望交付物,并对照来源检查输出。
npx skills add https://github.com/yashagarwal1994/dubai-dld-market-pricing-skill --skill dubai-dld-market-pricing开始前需了解
- 安装不包含模型订阅、API 额度或访问文件的权限。
- 仓库星数和安装量反映活跃情况,不等于独立质量或安全测试。
仓库原始内容
按源仓库内容呈现,保留标题、案例、代码、表格、链接以及原文引用的演示图片。
Dubai DLD Market Pricing
Overview
Use this skill to run repeatable, evidence-based pricing checks for Dubai communities using the public DLD transaction dataset. It is built for quick "is it cooling?" or "is it going up?" decisions based on recent closed transactions.
Workflow
- Pick the mode:
trends,latest, orproperty. - Run the analysis script against the official DLD endpoint.
- For
trends, compare recent windows against prior windows using medians. - For
latest, return newest sales across all of Dubai. - For
property, compare selected transaction versus prior comparable sales. - Report aggregate metrics with sample sizes, dates, and caveats when sparse.
Run Analysis
Use preset targets (Damac Hills 2, Emaar South, JLT, Reportage, Gemz Danube):
bash
scripts/dubai_dld_price_trends.py --preset all --days 90 --windows 14,30Use specific presets:
bash
scripts/dubai_dld_price_trends.py --preset damac-hills-2 --preset jlt --days 120Fetch latest city-wide sales:
bash
scripts/dubai_dld_price_trends.py --mode latest --latest-limit 25Inspect one transaction and check if price moved down/up:
bash
scripts/dubai_dld_price_trends.py \
--mode property \
--transaction-id "1-102-2026-21388" \
--property-area-tolerance-pct 5 \
--history-limit 20Use plain text area/society lookups:
bash
scripts/dubai_dld_price_trends.py --area "Jumeirah Village Circle" --area "Emaar South" --days 120Use listing URLs from Property Finder or Bayut:
bash
scripts/dubai_dld_price_trends.py \
--source-url "https://www.propertyfinder.ae/en/plp/buy/apartment-for-sale-dubai-jumeirah-lake-towers-jlt-12345678.html" \
--source-url "https://www.bayut.com/for-sale/apartments/dubai/jumeirah-lake-towers-jlt/" \
--days 120 --windows 14,30Use custom community matchers:
bash
scripts/dubai_dld_price_trends.py \
--target "business-bay=\\bbusiness\\s*bay\\b" \
--target "downtown=\\bdowntown\\s*dubai\\b" \
--days 90 --windows 14,30Persist outputs:
bash
scripts/dubai_dld_price_trends.py \
--preset all \
--output-json output/summary.json \
--output-markdown output/report.md \
--output-matches output/matches.jsonlInterpret Results
- Prefer
meter_sale_pricemedian as primary price signal. - Use
actual_worthmedian as secondary confirmation. - Treat outcomes with low counts (
n < 10) as directional only. - If 14-day and 30-day deltas disagree, classify as mixed/noisy.
- URL-derived targets rely on listing slug/query text; if matching is weak, add
--areaor explicit--target. - In
propertymode, if strict area matching is sparse, fallback uses broader same-profile comparables.
Privacy Rules
- Do not include private user data in prompts, outputs, or files.
- Keep outputs aggregate-only (counts, medians, deltas, windows).
- Avoid sharing raw personal records or non-essential row fields.
- Keep references to source endpoints and analysis dates for auditability.
Data Source
Read endpoint and field notes in references/dld-data-source.md.