walk-kcal
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| AMAP_KEY | No | Optional. Amap (Gaode) web-service key. Without it the server reads a bundled snapshot of pre-collected neighborhoods instead of geocoding a live address. | |
| MCD_MCP_TOKEN | No | Optional. Token for the official McDonald's China MCP. Without it, nutrition values come from a bundled snapshot of the official table. | |
| WALKKCAL_MCD_LIVE | No | Optional. Set to 1 to call the live McDonald's MCP instead of the snapshot. Requires MCD_MCP_TOKEN. | 0 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| walk_allowanceA | 卡路里额度核算。给定步行距离，算出可用的本餐额度。当用户问"走过去能吃多少""额度还剩多少"时使用。 |
| nearby_foodA | 附近门店 + 每道菜能不能吃下。跨数据源（麦当劳 MCP + 周边本地餐厅）。当用户问"附近有什么能吃的""现在吃什么""有什么推荐的"时使用。 ★ 用户若给出限定条件，填对应参数（不要自己过滤结果）： · "只看快餐/附近有什么快餐/想吃西餐/来点小吃" → industry · "不想走太远/就附近/500 米内" → radius_m=500 · "还有别的吗/换一批/再看看" → page 加 1 不带任何参数调用也能返回合理结果（默认半径 1500 米、第 1 批）。 返回值里的 next 字段是建议的下一步参数，可直接采用。 |
| what_can_i_eatA | 额度内能吃什么（跨数据源）。当用户问"我现在能点什么""有什么便宜（低卡）的"时使用。 用户给业态限定时填 industry；用户说"还有吗"时用 nearby_food 的 page 参数。 |
| explain_conceptB | 科普 NEAT / MET 的含义与依据，不消耗任何 MCP 调用。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
walk_allowance and explain_concept are clearly distinct, but nearby_food and what_can_i_eat overlap significantly for general 'what can I eat' queries. Descriptions try to separate them by context (nearby vs. within allowance), but the boundary remains fuzzy and could lead to misselection.
All tool names use snake_case, but the naming patterns are mixed: walk_allowance (noun phrase), nearby_food (adjective_noun), what_can_i_eat (question phrase), and explain_concept (verb_noun). This inconsistency makes the set less predictable despite being readable.
Four tools cover the core decision support loop: calculate allowance, discover nearby food, filter by allowance, and explain concepts. Each tool earns its place with no redundancy, and the count is well-suited to the server's narrow purpose.
The surface covers the main workflow: compute allowance, find nearby options, select within allowance, and educate on concepts. Minor gaps exist, such as no tool for detailed nutrition per dish or user preference tracking, but the core decision-making loop is complete.