Skip to main content
Glama

Deals picked for this person

recommend_promotions
Read-onlyIdempotent

Personal picks of current Singapore deals. Use for 'what deals would suit me?', 'any good deals for a student near NUS?', 'deals for my family this weekend', 'I'm visiting Singapore 3-7 Dec, staying near Bugis'. Pass what you already know from the conversation: who they are (life_stage, age_range), budget, interests, categories to skip, where they are (area or lat/lng), and/or ids of deals they liked. Each result has a short reason. Nothing is stored and no login is needed; don't ask the person questions only to fill these in.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "recommend_promotions"
  }
}

Input Schema

NameRequiredDescriptionDefault
latNoOptional: where they are (latitude), to rank nearby deals higher.
lngNoOptional: longitude, with lat.
areaNoInstead of lat/lng: an MRT area, mall or campus name, e.g. 'Tampines', 'Orchard', 'Jewel', 'NUS', 'NTU', 'SMU', 'Ngee Ann Poly'. Unknown names return an error; then pass lat/lng.
whenNoOnly deals usable then, Singapore time, reading each deal's hours (e.g. 'Weekdays 3pm-5pm'). 'dinner tonight' = intent dinner + when tonight.
limitNoHow many suggestions to return (1-30).
budgetNoOptional: low = around S$10 or less, freebies and 1-for-1s first; mid = everyday; high = treats (buffets, staycations) are fine.
intentNoWhat the person wants: breakfast, lunch, dinner, coffee (coffee, bubble tea, desserts), drinks, shopping, fun, money (cards & banks).
categoryNoKind of deal.
trip_endNoOptional YYYY-MM-DD: last day of the visit (with trip_start).
active_onNoOnly deals valid on this Singapore date (YYYY-MM-DD).
age_rangeNoOptional: the person's age range, if known. Used only when life_stage is not given (18_24 leans student, 55_plus leans senior); under_18 also sinks alcohol deals. Not stored.
interestsNoOptional: what they like. Category keys (food, grocery, shopping, fashion, beauty, health, electronics, home, entertainment, travel, finance, insurance, telco, car, property, education) or finer words (bubble_tea, coffee, desserts, fast_food, buffet, korean, japanese, drinks, movies, kids, fitness, staycation, gadgets). Matching deals rank first.
liked_idsNoOptional: ids of deals they liked or saved (from search results); finds similar ones.
radius_kmNoOptional, with a place: only deals within this many km (island-wide 'all outlets' deals still count).
life_stageNoOptional: who the deals are for, if you already know it from the conversation (never ask just for this). student, working, family (parents with kids), senior, tourist (visiting Singapore). Ranks fitting deals first, e.g. student = cheap food, bubble tea, student discounts first; hotels, cards and insurance later. tourist = deals a visitor can use (no Singapore bank card, residency or local scheme needed) first, attractions and Changi Airport deals up, and each deal gets tourist_fit {usable: yes|maybe|no, needs}. Not stored.
trip_startNoOptional YYYY-MM-DD (Singapore): first day of a visit or of the dates they care about. With trip_end, only deals valid on at least one of those days are returned.
exclude_categoriesNoOptional: categories to leave out entirely, e.g. ['travel', 'property'] for someone who said hotels don't interest them.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / intent / enum
      Previous value: -[
      -  "breakfast",
      -  "lunch",
      -  "dinner",
      -  "coffee",
      -  "drinks",
      -  "shopping",
      -  "fun",
      -  "money"
      -]New value: +[
      +  "breakfast",
      +  "lunch",
      +  "dinner",
      +  "coffee",
      +  "drinks",
      +  "birthday",
      +  "shopping",
      +  "fun",
      +  "money"
      +]
  2. Changed18 schema fields changed
    • addedInput schema / properties / active_on
      Added value: +{
      +  "description": "Only deals valid on this Singapore date (YYYY-MM-DD).",
      +  "type": "string"
      +}
    • addedInput schema / properties / age_range
      Added value: +{
      +  "description": "Optional: the person's age range, if known. Used only when life_stage is not given (18_24 leans student, 55_plus leans senior); under_18 also sinks alcohol deals. Not stored.",
      +  "enum": [
      +    "under_18",
      +    "18_24",
      +    "25_34",
      +    "35_44",
      +    "45_54",
      +    "55_plus"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / area
      Added value: +{
      +  "description": "Instead of lat/lng: an MRT area, mall or campus name, e.g. 'Tampines', 'Orchard', 'Jewel', 'NUS', 'NTU', 'SMU', 'Ngee Ann Poly'. Unknown names return an error; then pass lat/lng.",
      +  "type": "string"
      +}
    • addedInput schema / properties / budget
      Added value: +{
      +  "description": "Optional: low = around S$10 or less, freebies and 1-for-1s first; mid = everyday; high = treats (buffets, staycations) are fine.",
      +  "enum": [
      +    "low",
      +    "mid",
      +    "high"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / category
      Added value: +{
      +  "description": "Kind of deal.",
      +  "enum": [
      +    "food",
      +    "grocery",
      +    "shopping",
      +    "fashion",
      +    "beauty",
      +    "health",
      +    "electronics",
      +    "home",
      +    "entertainment",
      +    "travel",
      +    "finance",
      +    "insurance",
      +    "telco",
      +    "car",
      +    "property",
      +    "education"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / exclude_categories
      Added value: +{
      +  "description": "Optional: categories to leave out entirely, e.g. ['travel', 'property'] for someone who said hotels don't interest them.",
      +  "items": {
      +    "enum": [
      +      "food",
      +      "grocery",
      +      "shopping",
      +      "fashion",
      +      "beauty",
      +      "health",
      +      "electronics",
      +      "home",
      +      "entertainment",
      +      "travel",
      +      "finance",
      +      "insurance",
      +      "telco",
      +      "car",
      +      "property",
      +      "education"
      +    ],
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • addedInput schema / properties / intent
      Added value: +{
      +  "description": "What the person wants: breakfast, lunch, dinner, coffee (coffee, bubble tea, desserts), drinks, shopping, fun, money (cards & banks).",
      +  "enum": [
      +    "breakfast",
      +    "lunch",
      +    "dinner",
      +    "coffee",
      +    "drinks",
      +    "shopping",
      +    "fun",
      +    "money"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / interests
      Added value: +{
      +  "description": "Optional: what they like. Category keys (food, grocery, shopping, fashion, beauty, health, electronics, home, entertainment, travel, finance, insurance, telco, car, property, education) or finer words (bubble_tea, coffee, desserts, fast_food, buffet, korean, japanese, drinks, movies, kids, fitness, staycation, gadgets). Matching deals rank first.",
      +  "items": {
      +    "enum": [
      +      "food",
      +      "grocery",
      +      "shopping",
      +      "fashion",
      +      "beauty",
      +      "health",
      +      "electronics",
      +      "home",
      +      "entertainment",
      +      "travel",
      +      "finance",
      +      "insurance",
      +      "telco",
      +      "car",
      +      "property",
      +      "education",
      +      "bubble_tea",
      +      "coffee",
      +      "desserts",
      +      "fast_food",
      +      "buffet",
      +      "korean",
      +      "japanese",
      +      "drinks",
      +      "movies",
      +      "kids",
      +      "fitness",
      +      "staycation",
      +      "gadgets"
      +    ],
      +    "type": "string"
      +  },
      +  "maxItems": 12,
      +  "type": "array"
      +}
    • addedInput schema / properties / lat
      Added value: +{
      +  "description": "Optional: where they are (latitude), to rank nearby deals higher.",
      +  "type": "number"
      +}
    • addedInput schema / properties / life_stage
      Added value: +{
      +  "description": "Optional: who the deals are for, if you already know it from the conversation (never ask just for this). student, working, family (parents with kids), senior, tourist (visiting Singapore). Ranks fitting deals first, e.g. student = cheap food, bubble tea, student discounts first; hotels, cards and insurance later. tourist = deals a visitor can use (no Singapore bank card, residency or local scheme needed) first, attractions and Changi Airport deals up, and each deal gets tourist_fit {usable: yes|maybe|no, needs}. Not stored.",
      +  "enum": [
      +    "student",
      +    "working",
      +    "family",
      +    "senior",
      +    "tourist"
      +  ],
      +  "type": "string"
      +}
    • changedInput schema / properties / liked_ids / description
      Previous value: -"Ids from search results."New value: +"Optional: ids of deals they liked or saved (from search results); finds similar ones."
    • removedInput schema / properties / liked_ids / minItems
      Removed value: -1
    • addedInput schema / properties / lng
      Added value: +{
      +  "description": "Optional: longitude, with lat.",
      +  "type": "number"
      +}
    • addedInput schema / properties / radius_km
      Added value: +{
      +  "description": "Optional, with a place: only deals within this many km (island-wide 'all outlets' deals still count).",
      +  "type": "number"
      +}
    • addedInput schema / properties / trip_end
      Added value: +{
      +  "description": "Optional YYYY-MM-DD: last day of the visit (with trip_start).",
      +  "type": "string"
      +}
    • addedInput schema / properties / trip_start
      Added value: +{
      +  "description": "Optional YYYY-MM-DD (Singapore): first day of a visit or of the dates they care about. With trip_end, only deals valid on at least one of those days are returned.",
      +  "type": "string"
      +}
    • addedInput schema / properties / when
      Added value: +{
      +  "description": "Only deals usable then, Singapore time, reading each deal's hours (e.g. 'Weekdays 3pm-5pm'). 'dinner tonight' = intent dinner + when tonight.",
      +  "enum": [
      +    "now",
      +    "today",
      +    "tonight",
      +    "tomorrow",
      +    "weekend"
      +  ],
      +  "type": "string"
      +}
    • removedInput schema / required
      Removed value: -[
      -  "liked_ids"
      -]
  3. Changed1 schema field changed
    • addedInput schema / properties / limit / description
      Added value: +"How many suggestions to return (1-30)."
  4. First observed

TDQS

A4/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare read-only, idempotent, non-destructive, closed-world behavior, and the description adds genuinely new facts: nothing is stored, no login is required, and each result carries a short reason. The instruction not to interrogate the user purely to populate parameters is useful behavioral guidance an agent would otherwise get wrong.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Purpose is front-loaded in the first sentence, followed by trigger examples and then one parameter-grouping sentence. That sentence partly re-lists fields already documented in the schema, which is the only slack; otherwise there is no filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a zero-required, 17-parameter recommendation tool with no output schema, the definition covers intent, sourcing strategy, privacy, auth, and the shape of results (short reason, tourist_fit). It could say more about how ranking trades off competing signals, but nothing an agent needs to call it correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With 100% schema coverage the baseline is 3, but the description earns more by grouping the 17 optional inputs into a mental model (identity, budget, interests, exclusions, location, liked ids) and telling the agent to source them from conversation rather than asking. The life_stage/age_range interaction and tourist handling are described in the schema itself, so the description does not need to repeat them.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

"Personal picks of current Singapore deals" names a specific resource, a scope (current, Singapore) and a personalization angle that distinguishes it from a generic promotion search. It never names a sibling tool, so the agent must infer the boundary with search_promotions or nearby_promotions on its own.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Four concrete user-utterance examples ("what deals would suit me?", "student near NUS", "family this weekend", "visiting 3-7 Dec near Bugis") make the intended trigger condition very legible. There is no explicit when-not or named alternative, but the "pass what you already know" and "don't ask questions only to fill these in" lines bound the usage well.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources