Lizo AI MCP Server

Invoke this tool when the user wants to understand how a specific ZIP code or address performs relative to its broader metro area. Typical triggers: "how does [ZIP] compare to the MSA," "is this ZIP above or below metro average," "benchmark [ZIP] against the market," "how does this submarket stack up."

Before calling: confirm you have a ZIP code. The MSA can be inferred from the ZIP — do not require the user to name it explicitly.

After receiving data: synthesize into a short market read and present a side-by-side comparison — ZIP value vs. MSA average vs. 3-mile radius, 5-mile radius — for each key metric. Lead with the overall verdict: is this ZIP a stronger or weaker submarket than the metro average, and on which metrics specifically. Cover rent trajectory, occupancy trend, supply pipeline, and 3-year outlook. Conclude with an investment climate rating. Flag any metric where the gap is unusually large (positive or negative).

Invoke this tool when the user wants to get data or an analysis on a specific ZIP code. Typical triggers: "what's the market like in [ZIP]," "give me the KPIs for [ZIP code]," "show me data for 85281," "what are the rent and occupancy numbers in this ZIP."

Before calling: confirm you have a ZIP code. Do not call without one. If the user provides an address instead of a ZIP, extract or infer the ZIP before calling.

After receiving data: lead with the strongest market signal — whether the ZIP is outperforming, in-line, or underperforming vs. its MSA. Present rent, occupancy, and rent growth as the primary metrics. Highlight the 3-year outlook if available. Synthesize into a short market read. Ask if the user wants full data. Create the data in a table.

Invoke this tool when the user wants to compare two or more metro areas against each other. Typical triggers: "compare Phoenix and Atlanta," "which MSA is performing better," "show me Dallas vs. Denver," "I'm deciding between two markets — which one is stronger."

Before calling: confirm at least two MSA names. If the user provides city names instead of formal MSA names, map them (e.g., "Chicago" → "Chicago-Naperville-Elgin MSA"). If only one MSA is given, ask for the second before calling.

After receiving data: synthesize into a short market read and present a structured comparison table with one row per MSA and columns for each key metric. Declare a clear winner per metric. End with an overall summary of which MSA presents the stronger opportunity, framed around the user's stated investment goal if known.

Invoke this tool when the user wants to compare two or more specific ZIP codes side-by-side. Typical triggers: "compare 60614 vs 60657," "which ZIP is performing better," "show me these two submarkets," "60614, 60657, 60660 — rank them."

Before calling: confirm you have at least two 5-digit ZIPs. With a single ZIP, use get_zip_code_data or location_analysis instead.

After receiving data: present a structured comparison table with one row per ZIP and columns for each key metric. Declare a clear winner per metric. End with an overall summary of which ZIP presents the stronger opportunity, framed around the user's stated investment goal if known.

Invoke this tool when the user wants to search for or discover ZIP codes matching specific criteria — not when they already have a ZIP and want data on it. Typical triggers: "find ZIP codes in [city/state] with [characteristic]," "which ZIPs are near [area]," "what ZIP code is [neighborhood/address] in," "find me submarkets in [market] that match [criteria]."

Before calling: confirm the search criteria — geography (city, MSA, or state) and at least one characteristic to filter by (rent range, occupancy threshold, property density, etc.). If the user provides only a general location with no filter criteria, ask what they're looking for before calling. Do not use this tool if the user already has a specific ZIP — use get_zip_code_data instead.

After receiving data: present the matching ZIP codes as a ranked or grouped list. For each ZIP, surface the metric that made it match the search criteria. If the result set is large, highlight the top 5 and offer to show more.

Filterable fields: cap_rate, rent_growth, appreciation, total_population, population_growth, median_household_income, share_of_renters, median_rent, employment_rate, median_house_value, violent_crime_rate, state, city

Invoke this tool when the user wants to search for or rank MSAs by KPI criteria. Typical triggers: "which MSAs have strong rent growth," "find markets with low vacancy," "top metros for multifamily in Texas."

Before calling: confirm at least one filter criterion or geography. Do not use this if the user already named a specific MSA — use compare_msa instead.

After receiving data: present matching MSAs as a ranked list. For each MSA, surface the metric that drove the ranking. Offer to drill into a specific MSA with compare_msa.

Invoke this tool when the user wants to understand how a specific ZIP code performs relative to its broader metro area. Typical triggers: "how does [ZIP] compare to the MSA," "is this ZIP above or below metro average," "benchmark [ZIP] against the market," "how does this submarket stack up."

Before calling: confirm you have a ZIP code. The MSA is inferred automatically — do not require the user to name it explicitly.

After receiving data: synthesize into a short market read and present a side-by-side comparison — ZIP value vs. MSA average vs. 3-mile radius vs. 5-mile radius — for each key metric. Lead with the overall verdict: is this ZIP a stronger or weaker submarket than the metro average, and on which metrics specifically. Use the notable_gaps list to flag metrics where the ZIP diverges from its MSA by 15% or more — these are the most actionable signals.

Invoke this tool when the user wants to compare rent levels or rent growth across markets, ZIP codes, or property types. Typical triggers: "where are rents growing fastest," "compare rents in [A] vs. [B]," "show me rent trends across these ZIPs," "which property type is commanding higher rents."

Before calling: collect at least two comparison targets (ZIP codes or MSA codes). For a ranking ("top markets by rent growth"), use find_zip_code or find_msa instead — this tool is for direct head-to-head comparison.

After receiving data: lead with the highest and lowest performers (highest_rent_growth / lowest_rent_growth). Present effective rent (median_rent) and rent_growth side by side for each geography. Break down rent_by_unit_type (studio / 1BR / 2BR / 3BR) to show which bedroom type drives the spread. Flag any market where rent growth is accelerating vs. plateauing — this is the most actionable signal for investment decisions.

Invoke this tool when the user wants the full property record for a specific apartment by street address or asset name. Typical triggers: "show me data for [address]," "what's the asset class on [property]," "pull up [property name]," "what's the occupancy at [address]."

Before calling: confirm you have a full street address or asset name. A ZIP code alone is not sufficient — use get_zip_code_data instead.

After receiving data: lead with asset identification (asset_name, address, city/state/zip) and asset_class (A/B/C grade) alongside headline operating metrics. Then break down unit counts and rents by bedroom type (studio/1BR/2BR/3BR, including rent_psf), building details (built_year, built_year_effective, floors, total_units, management company), special_type, amenities, school_district, walk/transit/bike scores, transportation, and policy flags (is_mfh, is_furnished, is_pet_allowed). Flag any gap between the asset's class and its market position.

Invoke this tool when the user wants to search for apartment properties by location — city, ZIP, MSA, or full address. Typical triggers: "find properties in [market]," "which assets in [ZIP] are class A," "search for garden-style assets in [city]," "find me a comp for [property]."

Before calling: confirm at least a geography. Optionally filter by property_type ('multifamily') or numeric filters (total_units, built_year, floor, rent_1_br, rent_2_br, is_mfh). If the user has already named a specific address, use get_asset_data instead.

After receiving data: present matching assets as a ranked list. Each row carries the full property record: identification (asset_id, asset_name, address, lat/lon), unit counts and rents by bedroom (with rent_psf), building details (built_year, built_year_effective, floors, management company), asset_class (A/B/C grade), special_type, amenities, school_district, walk/transit/bike scores, transportation, and policy flags (is_mfh, is_furnished, is_pet_allowed). Surface asset_class up front when ranking — it's the most actionable signal.

Invoke this tool when the user wants to understand how a specific ZIP code or address performs relative to its broader metro area. Typical triggers: "how does [ZIP] compare to the MSA," "is this ZIP above or below metro average," "benchmark [ZIP] against the market," "how does this submarket stack up."

Before calling: confirm you have a ZIP code, a full street address, or a city/neighborhood description. The tool will geocode and extract the ZIP automatically.

After receiving data: synthesize into a short market read and present a side-by-side comparison — ZIP value vs. MSA average vs. 3-mile radius, 5-mile radius — for each key metric. Lead with the overall verdict: is this ZIP a stronger or weaker submarket than the metro average, and on which metrics specifically. Cover rent trajectory, occupancy trend, supply pipeline, and 3-year outlook. Conclude with an investment climate rating.

Invoke this tool when the user wants raw data for a specific ZIP code — not a comparative analysis. Typical triggers: "what's the market like in [ZIP]," "give me the KPIs for [ZIP code]," "show me data for 85281," "what are the rent and occupancy numbers in this ZIP."

Before calling: confirm you have a 5-digit ZIP code. If the user provides an address instead, extract or infer the ZIP first. For a market vs. metro benchmark, use location_analysis instead.

After receiving data: lead with the strongest market signal — whether the ZIP is outperforming, in-line, or underperforming. Present rent, occupancy, and rent growth as the primary metrics. Surface the 3-year outlook if available. Offer the full KPI table on request.

Invoke this tool when the user wants to compare two or more specific ZIP codes side-by-side. Typical triggers: "compare 60614 vs 60657," "which ZIP is performing better," "show me these two submarkets," "60614, 60657, 60660 — rank them."

Before calling: confirm you have at least two 5-digit ZIPs. With a single ZIP, use get_zip_code_data or location_analysis instead.

After receiving data: present a structured comparison table with one row per ZIP and columns for each key metric. Declare a clear winner per metric. End with an overall summary of which ZIP presents the stronger opportunity, framed around the user's stated investment goal if known.

Invoke this tool when the user wants to search for or discover ZIP codes matching specific criteria — not when they already have a ZIP and want data on it. Typical triggers: "find ZIP codes in [city/state] with [characteristic]," "screen submarkets with [criteria]," "top ZIPs by [KPI]."

Before calling: confirm the search criteria — geography (city, MSA, or state) and at least one characteristic to filter by. Do not use this tool if the user already has a specific ZIP — use get_zip_code_data instead.

After receiving data: present the matching ZIP codes as a ranked list. For each ZIP, surface the metric that made it match. If the result set is large, highlight the top 5 and offer to show more.

Filterable fields: cap_rate, rent_growth, appreciation, total_population, population_growth, median_household_income, share_of_renters, median_rent, employment_rate, median_house_value, violent_crime_rate, state, city

Invoke this tool when the user wants detailed year-over-year KPI data for a specific MSA. Typical triggers: "show me data for [MSA]," "give me the 5-year trajectory for [metro]," "what are the KPIs for the Phoenix MSA."

Before calling: confirm you have the MSA code. If the user gives a city or metro name, use find_msa first to discover the correct code.

After receiving data: present year-over-year trends across the key KPIs — cap rate, rent growth, occupancy, supply, population. Highlight inflection points and the 3-year outlook.

Invoke this tool when the user wants to search for or rank MSAs by KPI criteria. Typical triggers: "which MSAs have strong rent growth," "find markets with low vacancy," "top metros for multifamily in Texas."

Before calling: confirm at least one filter criterion or geography. Do not use this if the user already named a specific MSA — use compare_msa instead.

After receiving data: present matching MSAs as a ranked list. For each MSA, surface the metric that drove the ranking. Offer to drill into a specific MSA with compare_msa.

Invoke this tool when the user references a specific building or asset by address and wants its full property record. Typical triggers: "show me data for [address]," "what's the asset class on [property]," "pull up [property name]," "what's the occupancy at [address]."

Before calling: confirm you have a full street address with at least street number and street name. A ZIP alone is not sufficient — use get_zip_code_data instead.

After receiving data: lead with identification (asset_id, asset_name, address, city/state/zip, lat/lon) and asset_class (A/B/C grade). Then present unit counts and rents by bedroom type (with rent_psf), building details (built_year, built_year_effective, floors, total_units, management company), special_type, amenities, school_district, walk/transit/bike scores, transportation, and policy flags (is_mfh, is_furnished, is_pet_allowed).

Invoke this tool when the user wants to discover specific properties — not when they already know the address. Typical triggers: "find properties in [market]," "search for garden-style assets in [city]," "which properties in ZIP [X] have 100+ units," "find me a comp for [property]."

Before calling: confirm at least a geography (city, ZIP, or MSA) and optionally a property type or size filter. If the user has a specific address, use get_asset_data instead.

After receiving data: present matching assets as a list. Each carries the full property record: identification, lat/lon, unit counts and rents by bedroom (with rent_psf), building details (built_year, built_year_effective, floors, management company), asset_class (A/B/C grade), special_type, amenities, school_district, walk/transit/bike scores, transportation, and policy flags (is_mfh, is_furnished, is_pet_allowed).