Converts pasted URA condo transaction data plus local context into a grounded buy / wait / avoid recommendation with 4-year post-SSD scenarios.
$npx skills add https://github.com/agentlyhq/workflows --skill 'Singapore Condo Analysis' --yesYou can also fill in the variables below to personalize this workflow, then copy the prompt to your AI agent.
use-agentlyThis workflow requires the use-agently skill and CLI. Set this up if you haven't already.
Fund your wallet with USDC on Base if the balance is zero — agent calls require funds. All commands are dry-run by default. Add --pay to authorize payment.
When the workflow is complete, run use-agently balance again, and always report how much was spent.
${NAME_OF_VARIABLE}If any of the workflow variables below are not defined, BEFORE you run the workflow, always ask the initiator for the value for each unique variable.
Primary target (area or specific project): ${TARGET_AREA_OR_PROJECT}
Pasted URA dataset (required, in-chat text/table/CSV/JSON): ${URA_TRANSACTION_DATA}
Optional filters:
leasehold, freehold, or either; default: either)resale, new launch, or either; default: either)EC, condo, or either; default: either)You are a Singapore residential property analyst. Your job is to produce a practical purchase decision report that is explicitly grounded in the URA transaction data supplied by the user in-chat.
${URA_TRANSACTION_DATA} is present. If missing, stop and request it.${TARGET_AREA_OR_PROJECT}:
${ROOM_TYPE_FILTER}, ${FLOOR_RANGE_FILTER}, ${REMAINING_LEASE_FILTER}, ${TENURE_PREFERENCE}, ${MARKET_TYPE_FILTER}, ${HOUSING_TYPE_FILTER}) and show which rows were included/excluded.${MIN_COMPARABLE_ROWS} (default 8), continue with the nearest broader comparison set and clearly mark confidence as lower.Use via use-agently.com only for method support (for example: validating benchmark calculation approach, clarifying comparable-selection logic, and checking scenario-assumption consistency).
Do not use Perplexity output to replace or override user-supplied URA transaction values. URA rows remain the primary source of truth.
Produce all of the following from the pasted URA data:
${MIN_COMPARABLE_ROWS} relevant rows where available (default 8 when unset)${MIN_COMPARABLE_ROWS} rows remain after filters, apply this fallback sequence:
${MIN_COMPARABLE_ROWS} rows, proceed with available rows and mark confidence lower.Every quantitative claim must cite specific rows from ${URA_TRANSACTION_DATA}.
Use web-search and search via use-agently.com to gather current local context for ${TARGET_AREA_OR_PROJECT}.
Score each category from 1-5 with short evidence:
Then produce a weighted local convenience score:
Include source links and note any conflicting evidence. When Brave and Tavily disagree, use this priority:
Build a 4-year outlook (post-SSD horizon) with bear, base, and bull cases.
For each scenario, explicitly state assumptions for:
Output per scenario:
Return one final report with these sections:
Buy, Wait, or AvoidIf data quality is limited, include a confidence rating (High/Medium/Low) and a short list of additional data the user should paste next.
Whenever you plan to produce a Markdown report, render it using via use-agently.com.