Update to Advan Research REI

{
  "name": "foot-traffic-analysis",
  "description": "Step-by-step workflow for analyzing foot traffic at a single retail property using Advan REI MCP data. Use this skill whenever Reina is asked about a property's traffic performance, visit trends, dwell time, visit frequency, or general location performance. Trigger on phrases like \"how is [property] performing\", \"show me traffic for [location]\", \"what's the foot traffic at [address]\", \"how have visits trended\", \"is this center busy\", \"pull the traffic data for\", or any request that centers on understanding a specific property's visitor activity — even if the user doesn't say \"foot traffic\" explicitly. When in doubt, use this skill before answering.\n",
  "included_files": [],
  "skill_md_contents": "---\nname: foot-traffic-analysis\ndescription: >\n  Step-by-step workflow for analyzing foot traffic at a single retail property\n  using Advan REI MCP data. Use this skill whenever Reina is asked about a\n  property's traffic performance, visit trends, dwell time, visit frequency,\n  or general location performance. Trigger on phrases like \"how is [property]\n  performing\", \"show me traffic for [location]\", \"what's the foot traffic at\n  [address]\", \"how have visits trended\", \"is this center busy\", \"pull the\n  traffic data for\", or any request that centers on understanding a specific\n  property's visitor activity — even if the user doesn't say \"foot traffic\"\n  explicitly. When in doubt, use this skill before answering.\n---\n\n# Foot Traffic Analysis Workflow\n\nA single-property traffic analysis has four stages: resolve the property,\npull core metrics, benchmark against the market, and synthesize. The stages\nbuild on each other — don't skip ahead or collapse them into one big query.\n\n---\n\n## Stage 1: Resolve the Property\n\nProperties in the REI platform are identified by three fields: **ALI** (the\nprimary unique identifier), **location name**, and **company name**. If the\nuser has referenced any of these — or if they've been established earlier in\nthe conversation — use them directly and proceed to Stage 2.\n\nIf the property is ambiguous — a partial name, an address, or a loose\ndescription like \"the Target on Route 9\" — use `search` to find it. Present\nthe top results concisely and ask the user to confirm before pulling data.\nDon't guess and proceed; a misidentified property produces confidently wrong\nanalysis.\n\nOnce confirmed, carry the ALI, location name, and company name forward. You'll\nreuse them across every tool call in this workflow.\n\n---\n\n## Stage 2: Pull Core Metrics\n\nCall these three tools in order. Each one adds a layer to the story.\n\n**Date range:** Use whatever date range the user has selected in the UI, or\nwhatever period they've specified in their question. Only ask about the date\nrange if it's genuinely ambiguous from context.\n\n### 2a. Traffic Summary — the volume story\n\nCall `traffic-summary` with the resolved property identifiers.\n\nKey outputs to capture:\n- Total visit count for the period\n- Month-over-month and year-over-year trend\n- Any notable spikes or drops\n\n### 2b. Dwell Time — the engagement story\n\nCall `dwell-time` for the same property and date range.\n\nDwell time tells you whether visitors are spending meaningful time or just\npassing through. A high-traffic property with low dwell time often signals\nconvenience-oriented shopping or a pass-through layout; higher dwell suggests\ndestination behavior or a strong tenant mix.\n\n### 2c. Visit Frequency — the loyalty story\n\nCall `visit-frequency` for the same property and date range.\n\nVisit frequency distinguishes properties that draw the same loyal customers\nrepeatedly from those that rely on a constant stream of new visitors. A\ngrocery-anchored center will typically show higher frequency than a fashion\nor home goods property — keep that in mind when interpreting the result.\n\n---\n\n## Stage 3: Benchmark Against the Market\n\nA traffic number without context is just a number. Call `ranking` to place\nthe property within its submarket or category using the same date range.\n\nThe ranking result gives you where this property sits relative to peers and\nwhether it's gaining or losing ground. Weight this step based on what the\nuser actually asked — if they want to know if the property is strong, lead\nwith rank; if they're just asking how busy it is, one sentence is enough.\n\n---\n\n## Stage 4: Synthesize and Present\n\n### Search for real-world context\n\nBefore writing the narrative, do a targeted web search for anything that\ncould explain or color what the data is showing. You're looking for things\nlike: recent anchor tenant changes or closures, new competition that opened\nnearby, local economic conditions, a major employer moving in or out, weather\nevents or regional disruptions, renovation or construction activity, or any\nnews that a real estate professional would find relevant when reading these\nnumbers.\n\nYou don't need to find something every time — if a search turns up nothing\nuseful, move on. But when you do find something that connects, weave it into\nthe narrative naturally rather than citing it as a footnote. The goal is to\nhelp the user understand *why* the traffic looks the way it does, not just\n*what* it looks like.\n\nSearch examples that tend to be productive:\n- \"[Property name] [city] anchor tenant\"\n- \"[Retail chain] closures [year]\" if a relevant tenant is in the news\n- \"[City/submarket] retail market [year]\" for broader context\n- \"[Shopping center name] renovation OR redevelopment\"\n\n### Lead with the headline\n\nOpen with one sentence that answers the user's actual question. Don't make\nthem read a table to find the takeaway.\n\n> \"Riverside Commons drew roughly 1.2M visits over the period, up 8%\n> year-over-year, putting it in the top quartile of grocery-anchored centers\n> in the submarket — a trend that tracks with the Whole Foods renovation that\n> completed last spring.\"\n\n### Core metrics table\n\nAlways include a table. It's the reference point the user will return to.\n\n| Metric | Value | vs. Prior Year |\n|--------|-------|----------------|\n| Total visits | X | +/- Y% |\n| Avg. monthly visits | X | +/- Y% |\n| Median dwell time | X min | +/- Y% |\n| Visit frequency (avg. visits/visitor) | X | +/- Y% |\n| Submarket rank | #X of Y | ↑/↓ Z positions |\n\nPopulate only the rows where you have data. If a metric wasn't returned,\nomit the row rather than leaving it blank.\n\n### Trend narrative\n\nTwo to three sentences on what's driving the numbers. Look for:\n- Whether growth is consistent or concentrated in specific months\n- Any divergence between metrics (visits up but dwell down — could signal a\n  tenant change or anchoring issue)\n- Seasonal or external factors, especially anything surfaced from the web\n  search\n\nBe specific when the data supports it. If the data and context together tell\na clear story, tell it directly.\n\n### Cite data vintage\n\nEnd with the data vintage. Example:\n\n> *Data covers [date range], via Advan's mobile device panel.*\n\n---\n\n## Handling Gaps and Anomalies\n\n**No data returned:** Say so plainly and offer a path forward. Don't speculate\nabout why.\n\n> \"I'm not finding traffic data for that property — it may not be in our\n> coverage area. Want me to check a nearby comparable, or should I flag this\n> for the support team?\"\n\n**Partial data:** Lead with what's available, be explicit about what's\nmissing. Don't bury the gap.\n\n**Anomalous results:** If a number looks implausible — a 400% traffic spike,\na dwell time under 2 minutes for a grocery store — flag it before presenting\nit. The web search may turn up an explanation; if it does, note it. If it\ndoesn't, say so and recommend verifying.\n\n> \"These numbers look unusually high relative to the submarket. I didn't find\n> an obvious external explanation — I'd verify before relying on them. Want\n> me to pull a peer comparison?\"\n\n---\n\n## Follow-Ups\n\nIf there's a natural next step given what you found, mention it. This should\nfeel like something a colleague would say at the end of a conversation, not a\nmenu of options. Sometimes there's nothing to add — that's fine too.\n"
}