Dubai Wealth Index Real Estate

View saved version →

---
name: building-report
description: Produce a complete, sourced research note on one Dubai apartment building — prices, rents, gross yields, growth and evidence for every unit size — using the Dubai Wealth Index tools. Use when the user asks to "tell me about", assess, research, write up or do due diligence on a named Dubai building or tower, or wants a report, spreadsheet or chart about one.
---

# Report on one Dubai building

Explicit user instructions take priority over this workflow.

## 1. Confirm the building

1. Call `search` with the name the user gave.
2. Check `resolution`, score and locality before accepting a hit. A
   `matchedName` alone only names a matched alias; fuzzy candidates need
   confirmation. For an exact alias match, say so in the answer:
   "Vera Residences is registered as Vera Tower, Business Bay."
3. If two hits are close, or `get_building` returns `resolved.building.confidence`
   of `fuzzy`, ask the user which building they mean before going further.

## 2. Read the whole building once

Call `get_building` with `detail: "report"`. Do not call it again per bedroom
type — the report already covers every publishable one. It returns:

- `identity` — registered name, area, DLD project. Quote the registered name.
- `bedrooms[]` — for each publishable unit size: sale and rent medians, gross
  yield and its price basis, short-term and off-plan figures where published,
  `growth` with both comparison windows, the area benchmark and a 6-year trend.
- `growthWindows` — the dates and bases every growth figure compares.
- `suppressed[]` — each figure that did not clear its evidence gate, and why.
- `limitations[]` — the caveats that travel with the figures.

## 3. Check the evidence when it matters

If a headline rests on few sales (a count under about 10), or the user asks
how solid a figure is, call `get_building_transactions` for the same building
and say whether the median rests on many comparable deals or a few scattered
ones. Check `factsEdition` against `metricsEdition` before reconciling counts.
Keep the same bedroom type, cohort and window. Label pooled rents explicitly;
an empty building-level rent list does not mean the rent pool has no contracts.

## 4. Write it up

- Lead with the registered name, area and data date (`dataAsOf`).
- For each unit size: price, rent, gross yield, and the sample behind each.
- Growth: give the percentage with both windows, e.g. "+8.2% (median AED 2,040
  /sqft on 46 sales vs AED 1,885 on 51 in the prior 12 months)".
- List the suppressed figures as "not published: <reason>", never as zero.
- Close with the limitations that apply, and link `attribution` / the
  `canonicalUrl`.
- For a chart or spreadsheet, use the report's `trend` rows and per-bedroom
  figures as the data; do not interpolate missing years.

Never add a figure the tools did not return, and never describe a gross yield
as a net return.

View saved version →

---
name: check-a-claim
description: Test a claim about Dubai property returns — an agent's or developer's yield, price, rent or growth figure — against registered Dubai Land Department and Ejari records, using the Dubai Wealth Index tools. Use when the user quotes a figure for a Dubai building, area or community and asks whether it is true, realistic, typical or supported.
---

# Check a Dubai property claim

Explicit user instructions take priority over this workflow.

1. **Pin the claim down.** Which building or area, which unit size, which
   figure (gross or net yield, asking or achieved rent, launch or resale
   price), and over what period. If a detail is missing, ask for it or state
   the assumption you made.
2. **Find the registered figure.** `search`, then `get_building` (with the
   bedroom type) or `get_area` / `get_villa_community`. Confirm the building
   identity if the search matched another name.
3. **Compare like with like.**
   - Our yields are GROSS: registered new-contract rent over the median
     registered price in the same 12 months. Net is below gross only for the
     same income, price denominator and period with positive costs. A net yield
     on another purchase price is not directly comparable to the index gross
     yield. Identify the claim's actual basis; do not guess it.
   - A claimed short-term (Airbnb) return rests on an occupancy assumption;
     state ours from the tool result.
   - Prices are achieved, not asking.
4. **Show the evidence.** Quote the sample sizes. If the claim is about a small
   building or a thin grain, call `get_building_transactions` and describe the
   spread of actual deals.
   Preserve building versus pooled rent evidence and the returned data date.
   Link each canonical source. Service-charge budget units are the index's
   interpretation of the source categories. Use the building record's matched `afterServiceCharge`
   adjustment, which still excludes other operating and finance costs.
5. **Verdict.** Say whether registered records support, contradict or cannot
   test the claim, and by how much (percentage points for yields). Do not
   recommend buying or not buying.

View saved version →

---
name: check-asking-price
description: Judge a Dubai apartment's or villa's asking price against registered sales of similar units in the same building or villa community — fair range, verdict, confidence, total cash to buy, rent and net return, and what to confirm — using the Dubai Wealth Index tools. Use when the user has a listing, an offer or a price for a unit in a named Dubai building or a villa or townhouse in a named community and asks "is this a good price", "is it overpriced", "what should I offer", "what is it worth" or "is this a good deal", or pastes a property listing.
---

# Check a Dubai asking price

Explicit user instructions take priority over this workflow.

## 1. Pin the unit down

You need the building (or, for a villa or townhouse, the community), the
bedroom type, the size and the asking price.

- From a pasted listing, take the building name, bedrooms, built-up area and
  price. Convert sq m to sq ft (× 10.7639). Ignore "price per sq ft" claims in
  the listing; the tool computes its own. For a villa use the built-up area,
  not the plot, and note the phase or sub-community ("Palmera 2") if given.
- Missing size: run the check without it (the tool prices the building's
  typical unit and says so), then tell the user the answer moves with size.
- Missing price: run it without one for the fair range alone.
- Ready or off-plan: pass `status` only when the user says. Otherwise the
  tool reads the project register (for a villa, which side most recent sales
  were on).
- Service charge invoice, current tenancy rent, a direct developer purchase
  (`includeAgencyFee: false`) or a letting agent (`managementPct`): pass them
  when the user gives them. Do not invent them.

## 2. Resolve the place

Call `search` with the name. A marketing name ("Vera Residences") resolves to
the registered one ("Vera Tower"); say so. If two hits are close, ask which
one. A villa community hit (kind `villa`) is checked by community.

## 3. Run the check

Apartment: call `check_property` once with `area`, `building`, `bedroom`,
`sizeSqft` and `askingPrice`. Villa or townhouse: pass `community` instead of
area and building, and `project` for the phase when the user named one; the
villa bedroom types are 1br to 6br_plus. It returns:

- `headline` and `nextStep`: the answer in one sentence each.
- `verdict.position`: `below`, `lower` (fair, lower half), `upper` (fair, upper
  half) or `above` the range, with `outsideBy` / `outsideByPct` and
  `vsEstimatePct`.
- `range`: low, middle (`estimate`) and high, in AED and per sq ft.
- `confidence.grade` and `confidence.backtest`: how often that grade's middle
  landed within 10% of real sales.
- `comparables` and `setAside`: the sales used, and the ones left out with why.
- `buying`, `rent`, `holding`, `returns`: total cash, rent, running costs,
  gross yield, net yield and net return on cash.
- `offplan`: progress, planned handover and developer, when unfinished.
- Villas: `evidence.scope` (`project` when the named phase had enough sales,
  else `all`), `evidence.matchedProjects`, and `projects`, the phase names a
  `project` can match. If the phase was too thin, say the range is
  community-wide.
- `toConfirm` and `askTheSeller`: what the record cannot see.

If `range` is null there were fewer than three comparable sales. Say so, give
the published median from `get_building` (a villa: `get_villa_community`), and
do not produce a verdict.

## 4. Answer

Lead with the verdict and the numbers behind it:

> AED 2.35M is about AED 60k (2.6%) above recent sales of similar 2-beds in
> Executive Tower G. Seven sales since February put the fair range at
> AED 2.18M–2.29M, middle AED 2.25M (high confidence: in backtests a
> high-confidence middle landed within 10% of the real price 76% of the time).

Then, briefly:

1. What to offer: `nextStep`, in your words.
2. The money: total cash to buy, rent, net return on cash. If
   `returns.netYieldPct` is null, say no service charge is on file and give

Preview of the first 4,000 characters. Read full instructions →

View saved version →

---
name: cite-dubai-property-figures
description: Answer a question about what a Dubai apartment, building, area or villa community sells for, rents for or yields, using the Dubai Wealth Index tools, and quote each figure with its basis. Use when the user names a Dubai building, area or community, asks for gross rental yields, price growth or rankings, or wants to check a claim about Dubai property returns.
---

# Answer and cite a Dubai property figure

Explicit user instructions take priority over this workflow. If the user asks
for a different format or a narrower answer, follow the user.

## 1. Resolve the place

- If the user names a place, call `search` first. Use the hit's `kind` to pick
  the next tool: apartment buildings go to `get_building`, apartment areas to
  `get_area`, villa communities to `get_villa_community`.
- A `matchedName` identifies the alias a candidate matched, not proof of
  identity. Check `resolution`, score and locality first. For a confirmed exact
  match, tell the user which registered building it is before quoting: "Vera Residences is registered as Vera Tower".
- If `search` returns several plausible hits, or a tool's `resolved` block
  reports `confidence: "fuzzy"`, ask which one the user means rather than
  guessing.
- If nothing matches, say no matching published record was found. Try the
  rent purpose when relevant. Absence does not establish nonexistence or
  insufficient evidence; say below-gate only when the returned reason says so.

## 2. Pick the tool for the question

| Question | Tool |
| --- | --- |
| One building, one unit size | `get_building` with `bedroom` |
| One building, every unit size (a report) | `get_building` with `detail: "report"` |
| Every building in an area | `get_area` |
| A villa community | `get_villa_community` |
| Offices or shops (commercial, retail), one area or the best areas | `get_commercial` (`area` for one; `sort` to rank) |
| One off-plan or development project ("is it on track?", launch price, developer) | `get_project` (name is enough; area optional) |
| Off-plan market: best area, launch-to-resale uplift, live projects, late handovers | `get_offplan` (`area`, `view: "handovers"`) |
| A developer's delivery record, or the best developers | `get_developer` |
| Is the market rising, how far from the peak | `get_price_index` |
| "Which buildings…" against thresholds | `screen_buildings` |
| Two to four buildings side by side | `compare_buildings` |
| A published top list | `list_rankings`, then `get_ranking` |
| Short term, Airbnb, holiday home, short vs long let (area or city) | `get_ranking` `best-short-term-premium` with `area` — see the short-term-rental skill |
| Whether a median rests on enough sales | `get_building_transactions` |
| Citywide picture, choosing where to drill in | `get_city_overview` |
| How a figure is defined | `get_methodology` |
| The user's saved buildings | `list_favorites` (`since` for what changed) |
| What changed since an earlier edition | `whats_new` with `since` |
| Pages no typed tool covers (off-plan projects, developers) | `search`, then `fetch` |
| Any dubaiwealthindex.com URL a response links to | `fetch` — not a web browser |

Ask a follow-up only when a required input is missing, for example the bedroom
type when the user asks for "the yield" of a building that publishes several.

## 3. Quote every figure with its basis

For each number you report, include:

- the bedroom type,
- the unit (AED, AED per sqft, AED per year, percent),
- the period (usually the last 12 months),
- the data date (`dataAsOf`),
- the evidence count (number of sales or rent contracts).

Then apply these rules:

- Yields are **gross**, before service charge, agency fees, maintenance and
  vacant months. Never call one a net return.
- Rents are registered contracts, not asking rents. Prices are recorded
  transactions, not appraisals.
- Short-term rental figures rest on an occupancy **assumption**. State the
  assumption returned by the to

Preview of the first 4,000 characters. Read full instructions →

View saved version →

---
name: get-started
description: Set up Dubai Wealth Index after installation, check its connection when requested, and help the user try a first property question. Use for get started, onboarding, setup, connection checks, or questions about its data and capabilities.
---

# Get started with Dubai Wealth Index

Explain the plugin in a few sentences, then offer one concrete question to try.
Explicit user instructions take priority over this guidance.

## Choose the requested path

- For "set up", "get started", or "is it connected", call `whoami` once if
  available. Report only the returned scopes, data date and saved count.
  `mcp:read` confirms property-data access; `mcp:favorites` confirms saved-building
  access. A missing favorites scope does not prevent public property research.
  A null saved count is unavailable, not zero. Do not list saved buildings
  unless the user asks to see them.
- If tools are unavailable or authentication fails, explain that the connection
  is not verified and direct the user to connect or reconnect in the host.
  Do not claim the plugin is ready merely because this skill is present.
- For "what can it do", explain capabilities without an account lookup.
- If the user already supplied a property question, proceed with `search` and
  the relevant research tool; do not stop at suggesting that same question.
- Keep setup to a brief status and one useful first question. Tool availability
  and packaged skill availability are separate: successful tools do not prove
  the host loaded a skill, and a missing skill does not prove the server is down.

## What to say

- Dubai Wealth Index publishes **registered** Dubai property figures: median
  sale prices from Dubai Land Department transactions and median rents from
  Ejari rental contracts. They are recorded prices, not asking prices or
  appraisals.
- Apartments are published per building and per area. Villas and townhouses on
  plot titles are published per community, because the register records no
  building for a villa.
- Every figure carries its bedroom type, unit, period, data date and the number
  of transactions behind it.
- Every tool reads, except `request_info`, which sends an enquiry the user
  has confirmed to Dubai Wealth Index. Signing in also lets it read the
  buildings the user has saved on dubaiwealthindex.com; it cannot save or
  remove them.

## What it can answer

- One building or villa community: price, rent, gross yield, growth — or a
  full report on a building covering every unit size.
- An area: every published building, ordered by yield, price or growth.
- A screen across the city: buildings above a yield floor or inside a price band.
- A side-by-side comparison of two to four buildings on the same bedroom type.
- A published ranking, or the individual transactions behind a median.
- Offices and shops: achieved prices, rents, yields, liquidity and leasing
  demand by area.
- Off-plan: any registered project's progress, planned and recorded dates,
  launch prices and developer; which areas' launches resold highest; late
  handovers; developers ranked on delivery; the citywide price index.
- Short-term rentals: which buildings in an area out-earn the registered lease
  on Airbnb, the break-even occupancy, and short- versus long-let yield.
- Whether an agent's or developer's claimed yield holds up against registered
  records.
- The user's saved buildings, and what changed in them since an earlier edition.
- An enquiry to a licensed agent about a property, sent only after the user
  confirms it (`request_info`).

- Registered apartment sales activity: counts, transaction values, price ranges
  and evidenced year-over-year change by area and bedroom.
- Pending project supply: planned dates, progress and known units, with source
  dates, overdue projects and missing-data coverage kept explicit.

## What it cannot do

Say so plainly if asked: it holds no listings, cannot book a viewing itself
(`request_info` passes the reque

Preview of the first 4,000 characters. Read full instructions →

View saved version →

---
name: request-info
description: Put the user in touch with a licensed Dubai real-estate agent about a building, off-plan project or villa community — available units, prices, payment plans, a viewing or an offer — using the Dubai Wealth Index request_info tool. Use when the user wants to buy, rent, view, make an offer, or talk to an agent about a Dubai property.
---

# Request info from an agent

Explicit user instructions take priority over this workflow.

1. **Pin the property.** Use the page url from an earlier result if you have
   one. Otherwise `search` the name and confirm the match with the user when
   it is fuzzy or there are several.
2. **Ask what they want, briefly.** Unit type, buy or rent, and anything that
   matters to them (budget, timing). Write it as the
   `message` in their words. Don't interrogate: the property alone is enough.
3. **Ask how to reach them.** The account's name and email are always sent.
   Ask whether they also want a call or WhatsApp, and on which number.
4. **Confirm, then send.** Say what will be sent and to whom (Dubai Wealth
   Index, which replies and introduces a licensed agent, and may pass the
   enquiry to the licensed brokerage it works with), get a yes, then call
   `request_info` with
   `consent: true`. Never call it on your own initiative.
5. **Report back.** Give the reference, who it went to (`sentTo`), and the
   WhatsApp link if one is returned. Follow the result's `next` line.

If `request_info` returns `received: false`, fix the named fields with the
user and call again. If it returns `found: false`, offer its suggestions or
send it as a general enquiry without `property`.

View saved version →

---
name: shortlist-and-compare
description: Build a shortlist of Dubai buildings from registered prices, yields and evidence thresholds, then compare the finalists side by side, using the Dubai Wealth Index tools. Use when the user asks which Dubai buildings or areas meet a budget, yield or rent target, wants the best buildings in an area for rental yield, or wants two to four named buildings compared.
---

# Shortlist and compare Dubai buildings

Explicit user instructions take priority over this workflow.

Resolve named buildings, areas and communities with `search` first. Confirm
ambiguous candidates before comparing or filtering on their identities.

## Shortlist

- Thresholds the user states (budget, minimum gross yield, bedroom type,
  minimum number of sales, areas) go to `screen_buildings`. Set `minEvidence`
  to at least 10 when the user wants robust figures, and say what you used.
  This threshold counts sales, not rent contracts. Inspect rent evidence
  separately; use `buildingRentOnly` when building-specific rents are requested.
  Keep explicit user bounds unchanged, and disclose the result limit. A
  truncated result is not an exhaustive shortlist.
- "Best yields in <area>" without thresholds: `get_area` sorted by yield, or a
  published ranking (`list_rankings`, then `get_ranking` with the area).
  A ranking lists outliers and needs ten qualifying buildings; say so.
- "Best for short term / Airbnb in <area>": `get_ranking` with
  `best-short-term-premium` and the area — it lists every qualifying building
  with the lease comparison. Follow the short-term-rental skill.
- Say plainly that thresholds apply to **registered medians**, not to
  apartments currently for sale, and that buildings below the publication gate
  cannot appear.

## Compare

- Two to four apartment buildings: `compare_buildings` on one bedroom type.
  It returns only the unit sizes every building publishes — keep that grain,
  and never compare one building's 1-bed with another's 2-bed.
- Villa communities: call `get_villa_community` once per community, same
  purpose and bedroom count.
- Areas: `get_area` per area on the same purpose and bedroom type.
- Offices or shops: `get_commercial` without an area ranks every commercial
  area (by yield, sales, price, demand or liquidity); `assetClass` picks
  offices or shops. Commercial has no building grain — say so if asked.

## Optional market context

When the question calls for activity or supply context, use `get_market_activity`
on matching bedroom and ready/off-plan cohorts, or `get_supply_pipeline` for
pending projects. Keep source dates and both growth windows. Activity is not
sale probability or days on market; planned dates are not forecasts. Show a
brief summary first and offer detailed evidence only when useful.

## Report

A table works best: building, area, median price, price per sq ft, median rent,
gross yield, sales and contracts behind them, data date. A difference between
yields is in percentage points. Link each building's `canonicalUrl`.

View saved version →

---
name: short-term-rental
description: Answer Dubai short-term rental questions — Airbnb, holiday home, short let versus long let, rental arbitrage, which buildings or bedroom types work best for short term in an area — using the Dubai Wealth Index tools. Use when the user asks what is best or most profitable for short term in a Dubai area or building, whether to short-let or lease, or about nightly rates, occupancy or break-even.
---

# Short-term rentals in Dubai

Explicit user instructions take priority over this workflow. Every
short-term figure the website shows is returned by these tools, and `fetch`
reads any dubaiwealthindex.com page a response links to.

## 1. Pick the call

| Question | Call |
| --- | --- |
| "What is best for short term in <area>?" | `get_ranking` with `ranking: "best-short-term-premium"`, `area` |
| Short let vs long lease in an area (or arbitrage: rent then sublet) | same, `sort: "premium"` (the default with an area) |
| Buy to short-let: return on the purchase price | same, `sort: "yield"` |
| One bedroom type ("studios in JVC") | same, add `bedroom` |
| Where in Dubai | `get_ranking` without `area`; `list_rankings` lists the areas that publish |
| One building | `get_building` with `detail: "report"`: each bedroom's `shortLet` block |

Pass the area as the user wrote it; names and slugs both resolve. One call
answers the area question: do not follow it with `get_building` per row — each
row already carries the figures below.

## 2. Read the row

- `medianAdr` nightly rate, with `listingCount` and `rateSource`
  (`building` = sampled in that building, `vicinity` = nearby listings).
- `netRevenue` a year after platform fee, cleaning and utilities.
- `medianRent` the registered annual lease; `surplusAfterLease` the gap in AED;
  `premiumVsLease` the ratio (1.3 = 30% more than leasing).
- `breakEvenOccupancyPct` the occupancy at which short letting only matches
  the lease — the most useful risk figure: the lower, the more margin.
- `netYieldPct` beside `longLetGrossYieldPct`: the same purchase price under
  both strategies, so the gap is in percentage points.

## 3. Answer

- Lead with the direct answer: the buildings (and bedroom types) that come
  out ahead, how far ahead of the lease, and the break-even occupancy.
- If the user asked "short or long", say which wins in this model and by how
  much, then the break-even as the condition for it.
- Give `areaMedians` for context, and say the list is every qualifying
  building in the area: only rows earning more than the lease are listed, so
  a building missing from it does not beat its lease in this model.
- Before any short-term figure, state the assumptions from the response
  notes: occupancy is an assumption (not a measurement), nightly rates are
  public asking rates (not bookings), and the unit is self-managed —
  professional management typically takes 15–25% of revenue on top.
- Holiday-home letting in Dubai needs a DET permit; mention it once when the
  user is deciding whether to do it. Do not give investment advice.
- Link `canonicalUrl` and each row's `url`, with `dataAsOf`.

If the area returns `below-gate`, no building there has both a nightly-rate
sample and a registered lease beating it — say so, then offer the citywide
list or the area's long-let yields (`get_area`), as the suggestions name.

View saved version →

---
name: track-saved-buildings
description: Review the Dubai buildings saved on the user's Dubai Wealth Index account and what changed in them since an earlier data edition, using the Dubai Wealth Index tools. Use when the user asks about their saved, favourite or watched buildings, their shortlist, or what moved in the index since they last looked.
---

# Track saved buildings

Explicit user instructions take priority over this workflow.

1. Call `list_favorites`. It lists saved buildings with current figures and
   names any saved building below the publication gate in `belowGate` — report
   those as "no current figure (too few registered transactions this
   edition)", never as deleted.
2. For "what changed": call `whats_new` to see the editions, then
   `list_favorites` with `since` set to the `snapshotDate` of the edition the user means (the
   previous one if unspecified). Each change gives both editions' median price,
   rent, gross yield and counts, with the move in percent (prices, rents) and
   percentage points (yield). Changes can reflect moving windows, late registrations or corrections;
   do not infer the cause from the edition list. Distinguish `snapshotDate`
   (edition identity) from `dataAsOf` (source cutoff), which can differ.
3. For buildings not on the account, `whats_new` with `since` and `buildings`
   gives the same comparison.
   If no buildings are saved, say so and link https://dubaiwealthindex.com/account. Never
   invent a previous edition or call an unavailable figure zero. Cite record
   URLs with the bedroom type, evidence counts and both dates.
4. No tool saves or removes a building: to add or remove one, point the user
   to dubaiwealthindex.com. If `list_favorites` reports a missing scope, ask
   the user to reconnect and allow saved buildings.