parsley-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PARSLEY_API_TOKEN | Yes | Your Parsley API token for authentication. |
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 | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| configure_tokenB | Set the Parsley API bearer token for this session |
| list_menu_itemsA | List menu items (recipes, subrecipes, ingredients). Returns up to 20 with {items, total, truncated}; each item has id, name, itemNumber, tags, type. If truncated, prefer search_menu_items or narrow with syncTag/type. |
| search_menu_itemsA | Fuzzy-ranked search of menu items over name/itemNumber/tags (typo-tolerant, multi-token). Returns {items, total, offset, truncated}, best matches first. Default page size 20 (max 50); use offset to page through ranked results. |
| get_menu_itemB | Get menu item details: description, nutrition, allergens, photo. |
| list_menusA | List menus. Returns up to 20 with {items, total, truncated}; each item has id, name. If truncated, prefer search_menus. |
| search_menusA | Fuzzy-ranked search of menus by name (typo-tolerant, multi-token). Returns {items, total, offset, truncated}, best matches first. Default page size 20 (max 50); use offset to page through ranked results. |
| get_menuB | Get menu with sections, stations, and items. |
| get_recipeB | Get recipe: steps, ingredients, sub-recipes, nutrition, cost. |
| list_ingredientsA | List ingredients. Returns up to 20 with {items, total, truncated}; each item has id, name, itemNumber, salable. If truncated, prefer search_ingredients or filter by salable. |
| search_ingredientsA | Fuzzy-ranked search of ingredients over name/itemNumber (typo-tolerant, multi-token). Returns {items, total, offset, truncated}, best matches first. Default page size 20 (max 50); use offset to page through ranked results. |
| get_ingredientB | Get ingredient: conversions, supply options, preparations. |
| list_eventsC | List events in a date range. |
| get_eventC | Get event with line items. |
| list_serving_stationsC | List serving stations. |
| list_chef_tagsC | List chef tags. |
| list_chef_usersB | List chef users. |
| get_access_tokenB | Get CloudFront access token for CDN. |
| clear_cacheA | Clear the cached Parsley API responses for this token. GET responses are cached for 24 hours; call this if you suspect data is stale. Ask the user for permission before calling, since subsequent reads will refetch from the API. |
| get_commissary_reportC | Commissary transaction report as CSV. |
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 19 tools
Each tool targets a distinct resource+action, and the list-vs-search pairs (list_menus/search_menus, list_ingredients/search_ingredients, list_menu_items/search_menu_items) are clearly differentiated by their descriptions. The only mild overlap is between get_menu, get_menu_item, and get_recipe, but these map to genuinely different entities so an agent can select correctly.
Every tool follows a clean snake_case verb_noun pattern (list_/search_/get_/clear_/configure_). The verbs are used consistently: list_* for enumeration, search_* for fuzzy query, get_* for single-entity fetch, matching the description semantics.
19 tools is at the higher end but justified given the domain spans menus, recipes, ingredients, menu items, events, plus utility/auth tools (configure_token, clear_cache, get_access_token). Nothing feels redundant given the list/search/detail triad per entity.
The surface covers read operations comprehensively across all entities (list, search, get for menus/ingredients/menu items, plus recipes, events, stations, chef tags/users, report). There are no write/update/delete operations, but the server appears intentionally read-only (24h GET cache), and only minor detail getters (e.g. serving station, chef user) are absent.