MindDory Brain is an MCP server published by com.minddory, hosted at api.minddory.com. When last checked on 2026-11-22 it was working: it answers and lists its tools. It declares 21 tools, including get_user_profile, get_system_instructions, get_minddory_guide, get_known_words, get_queue.
Free: turn your AI chats into spaced-repetition vocabulary. 13 tools, reads and writes.
Nothing else in the archive shares these values.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
Inferred from the host name and headers: what a server runs on, not who runs it.
Profile snapshot: CEFR level, target/source languages, due card count, weak words, recent lookups. Pass `lang` to scope the snapshot to one learning language (for users learning several); omit it for the user's primary language.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Optional ISO or BCP-47 tag (en, en-us, pt-br). Scopes the whole profile - level, due count, weak words - to that learning language. Omit for the user's primary language. |
{
"name": "get_user_profile",
"title": "Get user profile",
"description": "Profile snapshot: CEFR level, target/source languages, due card count, weak words, recent lookups. Pass `lang` to scope the snapshot to one learning language (for users learning several); omit it for the user's primary language.",
"inputSchema": {
"type": "object",
"properties": {
"lang": {
"type": "string",
"description": "Optional ISO or BCP-47 tag (en, en-us, pt-br). Scopes the whole profile - level, due count, weak words - to that learning language. Omit for the user's primary language."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get user profile",
"readOnlyHint": true,
"openWorldHint": false,
"destructiveHint": false
}
}Returns the MindDory setup for this conversation: the user's CEFR level, target and source languages (with regional variant), due-card count, weak words, and the capture guidance for their language. `lang` scopes the result to one practice language (ISO or BCP-47, e.g. en-gb); without it the user's primary learning language is used.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Optional ISO or BCP-47 tag of the language the user wants to practice in this conversation (en, en-us, de, pt-br). Defaults to the user's primary learning language. |
{
"name": "get_system_instructions",
"title": "Get session instructions",
"description": "Returns the MindDory setup for this conversation: the user's CEFR level, target and source languages (with regional variant), due-card count, weak words, and the capture guidance for their language. `lang` scopes the result to one practice language (ISO or BCP-47, e.g. en-gb); without it the user's primary learning language is used.",
"inputSchema": {
"type": "object",
"properties": {
"lang": {
"type": "string",
"description": "Optional ISO or BCP-47 tag of the language the user wants to practice in this conversation (en, en-us, de, pt-br). Defaults to the user's primary learning language."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get session instructions",
"readOnlyHint": true,
"openWorldHint": false,
"destructiveHint": false
}
}Returns a short guide to MindDory as used from this chat: what it does, which tools it offers, the settings the user can change in the app (which actions are allowed, default folder for new words, hidden folders, change history with Undo), situations where it is useful, and the user's own current settings. Read-only, takes no arguments.
No parameters.
{
"name": "get_minddory_guide",
"title": "Get MindDory guide",
"description": "Returns a short guide to MindDory as used from this chat: what it does, which tools it offers, the settings the user can change in the app (which actions are allowed, default folder for new words, hidden folders, change history with Undo), situations where it is useful, and the user's own current settings. Read-only, takes no arguments.",
"inputSchema": {
"type": "object",
"properties": {},
"additionalProperties": false
},
"annotations": {
"title": "Get MindDory guide",
"readOnlyHint": true,
"openWorldHint": false,
"destructiveHint": false
}
}Words the user has verified known via flashcard practice. Paginated keyset on flashcards.id.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Language tag of the words: base ISO with an optional region (en, en-us, pt-br). Omit it and the user's own primary learning language is used. |
| limit | integer | |
| cursor | string |
{
"name": "get_known_words",
"title": "Get known words",
"description": "Words the user has verified known via flashcard practice. Paginated keyset on flashcards.id.",
"inputSchema": {
"type": "object",
"properties": {
"lang": {
"type": "string",
"description": "Language tag of the words: base ISO with an optional region (en, en-us, pt-br). Omit it and the user's own primary learning language is used."
},
"limit": {
"type": "integer",
"maximum": 500,
"minimum": 1
},
"cursor": {
"type": "string"
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get known words",
"readOnlyHint": true,
"openWorldHint": false,
"destructiveHint": false
}
}Cards due now and due within the next 24 hours.
| Parameter | Type | Description |
|---|---|---|
| max | integer | |
| lang | string | Language tag of the queue: base ISO with an optional region (en, en-us, pt-br). Omit it and the user's own primary learning language is used. |
{
"name": "get_queue",
"title": "Get review queue",
"description": "Cards due now and due within the next 24 hours.",
"inputSchema": {
"type": "object",
"properties": {
"max": {
"type": "integer",
"maximum": 100,
"minimum": 1
},
"lang": {
"type": "string",
"description": "Language tag of the queue: base ISO with an optional region (en, en-us, pt-br). Omit it and the user's own primary learning language is used."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get review queue",
"readOnlyHint": true,
"openWorldHint": false,
"destructiveHint": false
}
}Single card detail by word (case-insensitive). Returns translation, mastery, and last 10 events.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Language tag of the word: base ISO with an optional region (en, en-us, pt-br). Omit it and the user's own primary learning language is used. |
| word * | string |
{
"name": "get_card",
"title": "Get card detail",
"description": "Single card detail by word (case-insensitive). Returns translation, mastery, and last 10 events.",
"inputSchema": {
"type": "object",
"required": [
"word"
],
"properties": {
"lang": {
"type": "string",
"description": "Language tag of the word: base ISO with an optional region (en, en-us, pt-br). Omit it and the user's own primary learning language is used."
},
"word": {
"type": "string",
"maxLength": 200,
"minLength": 1
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get card detail",
"readOnlyHint": true,
"openWorldHint": false,
"destructiveHint": false
}
}Event log slice with optional surface filter and keyset pagination on answers.id.
| Parameter | Type | Description |
|---|---|---|
| limit | integer | |
| cursor | string | |
| surface | string | e.g. 'claude_mcp', 'app', 'cursor_mcp' |
{
"name": "get_recent_activity",
"title": "Get recent activity",
"description": "Event log slice with optional surface filter and keyset pagination on answers.id.",
"inputSchema": {
"type": "object",
"properties": {
"limit": {
"type": "integer",
"maximum": 100,
"minimum": 1
},
"cursor": {
"type": "string"
},
"surface": {
"type": "string",
"description": "e.g. 'claude_mcp', 'app', 'cursor_mcp'"
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get recent activity",
"readOnlyHint": true,
"openWorldHint": false,
"destructiveHint": false
}
}Append a generic interaction event to the answers log. Event types: a lookup, a word discussed with the AI, or a word read in context.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Language tag of the word: base ISO with an optional region (en, en-us, pt-br). Matching is on the base code, so any regional variant matches. Omit it and the user's own primary learning language is used. |
| type * | string | one of ["lookup","discussed","read_in_context"] |
| word * | string | |
| metadata | object |
{
"name": "log_interaction",
"title": "Log interaction",
"description": "Append a generic interaction event to the answers log. Event types: a lookup, a word discussed with the AI, or a word read in context.",
"inputSchema": {
"type": "object",
"required": [
"type",
"word"
],
"properties": {
"lang": {
"type": "string",
"description": "Language tag of the word: base ISO with an optional region (en, en-us, pt-br). Matching is on the base code, so any regional variant matches. Omit it and the user's own primary learning language is used."
},
"type": {
"enum": [
"lookup",
"discussed",
"read_in_context"
],
"type": "string"
},
"word": {
"type": "string",
"maxLength": 200,
"minLength": 1
},
"metadata": {
"type": "object",
"additionalProperties": true
}
},
"additionalProperties": false
},
"annotations": {
"title": "Log interaction",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": false
}
}Confidence-weighted spaced-repetition boost when the user has used a word correctly: the card moves further out in the review schedule. Logs an answer row even if no flashcard exists.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Language tag of the word: base ISO with an optional region (en, en-us, pt-br). Matching is on the base code, so any regional variant matches. Omit it and the user's own primary learning language is used. |
| word * | string | |
| source | string | |
| context | string | |
| confidence * | string | one of ["high","medium","low"] |
{
"name": "mark_demonstrated",
"title": "Mark word demonstrated",
"description": "Confidence-weighted spaced-repetition boost when the user has used a word correctly: the card moves further out in the review schedule. Logs an answer row even if no flashcard exists.",
"inputSchema": {
"type": "object",
"required": [
"word",
"confidence"
],
"properties": {
"lang": {
"type": "string",
"description": "Language tag of the word: base ISO with an optional region (en, en-us, pt-br). Matching is on the base code, so any regional variant matches. Omit it and the user's own primary learning language is used."
},
"word": {
"type": "string",
"maxLength": 200,
"minLength": 1
},
"source": {
"type": "string",
"maxLength": 100
},
"context": {
"type": "string",
"maxLength": 2000
},
"confidence": {
"enum": [
"high",
"medium",
"low"
],
"type": "string"
}
},
"additionalProperties": false
},
"annotations": {
"title": "Mark word demonstrated",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": false
}
}Spaced-repetition degrade for a word the user just got wrong: the card comes back sooner. ease_factor drops, interval resets, repetitions reset.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Language tag of the word: base ISO with an optional region (en, en-us, pt-br). Matching is on the base code, so any regional variant matches. Omit it and the user's own primary learning language is used. |
| word * | string | |
| source | string | |
| context | string |
{
"name": "mark_struggled",
"title": "Mark word struggled",
"description": "Spaced-repetition degrade for a word the user just got wrong: the card comes back sooner. ease_factor drops, interval resets, repetitions reset.",
"inputSchema": {
"type": "object",
"required": [
"word"
],
"properties": {
"lang": {
"type": "string",
"description": "Language tag of the word: base ISO with an optional region (en, en-us, pt-br). Matching is on the base code, so any regional variant matches. Omit it and the user's own primary learning language is used."
},
"word": {
"type": "string",
"maxLength": 200,
"minLength": 1
},
"source": {
"type": "string",
"maxLength": 100
},
"context": {
"type": "string",
"maxLength": 2000
}
},
"additionalProperties": false
},
"annotations": {
"title": "Mark word struggled",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": false
}
}Saves one or more target-language words or phrases from the conversation to the user's MindDory deck. A new word becomes a flashcard in the user's default folder when it is in the word's language, otherwise in a chat folder for that language; a word already in the deck gets the new sentence recorded as another encounter. `lang` (ISO code of the word's own language) is required and the call is rejected without it; the target language is never guessed. `words` takes several items at once; `gloss` is a short source-language translation and `context` the sentence the word appeared in. Items without a letter (numbers, punctuation), items shorter than 3 characters (CJK and Hangul exempt) and common stop words in English, German, Spanish, French and Czech are skipped. `source_lang` defaults to the user's deck.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Required. ISO code of the captured word's own language (en, de, es, zh, ...). The call is rejected without it. |
| word | string | Single-word capture (legacy); `words` takes several. |
| gloss | string | Short translation in the source language. |
| words | array | Batch capture. Each entry is {word, gloss?, context?}. |
| context | string | The sentence where the word appeared. |
| folder_id | integer | Optional: id of the folder new cards go into (from list_folders). It must be a folder in the same language as `lang`. Omit it and the user's default folder is used when it is in the same language, otherwise a chat folder for that language. Words already in the deck stay where they are. |
| source_lang | string |
Records a grammar mistake from a target-language sentence the user wrote, as a grammar point in their MindDory deck (shown on the Grammar Patterns page at app.minddory.com/grammar). Takes the original text, the correction and an error_type: tense, agreement, word_choice, word_order, article, preposition, spelling, pluralization, conjugation, gender or other. One call records one mistake. `lang` takes the full BCP-47 variant (en-gb, pt-br) so the mistake is filed under the user's regional variant.
| Parameter | Type | Description |
|---|---|---|
| lang | string | ISO target language code. |
| user_text * | string | The exact erroneous sentence/phrase the user wrote. |
| correction * | string | The corrected sentence/phrase. |
| error_type | string | One of: tense, agreement, word_choice, word_order, article, preposition, spelling, pluralization, conjugation, gender, other. |
| explanation | string | Short rationale (1-2 sentences). Optional. |
{
"name": "capture_grammar_mistake",
"title": "Capture grammar mistake",
"description": "Records a grammar mistake from a target-language sentence the user wrote, as a grammar point in their MindDory deck (shown on the Grammar Patterns page at app.minddory.com/grammar). Takes the original text, the correction and an error_type: tense, agreement, word_choice, word_order, article, preposition, spelling, pluralization, conjugation, gender or other. One call records one mistake. `lang` takes the full BCP-47 variant (en-gb, pt-br) so the mistake is filed under the user's regional variant.",
"inputSchema": {
"type": "object",
"required": [
"user_text",
"correction"
],
"properties": {
"lang": {
"type": "string",
"description": "ISO target language code."
},
"user_text": {
"type": "string",
"maxLength": 2000,
"minLength": 1,
"description": "The exact erroneous sentence/phrase the user wrote."
},
"correction": {
"type": "string",
"maxLength": 2000,
"minLength": 1,
"description": "The corrected sentence/phrase."
},
"error_type": {
"type": "string",
"maxLength": 40,
"description": "One of: tense, agreement, word_choice, word_order, article, preposition, spelling, pluralization, conjugation, gender, other."
},
"explanation": {
"type": "string",
"maxLength": 1000,
"description": "Short rationale (1-2 sentences). Optional."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Capture grammar mistake",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": false
}
}Get the user's most actively encountered target-language words (from past capture_word + log_interaction events), ranked by frequency over a lookback window. These are the "frontier" words the user keeps meeting.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Language tag (en, en-us, pt-br). Matching is on the base code, so a variant finds encounters logged under any sibling. Omit it to include ALL languages - unlike the deck tools, this one is a cross-language frequency view. |
| limit | integer | Default 20. |
| lookback_days | integer | Default 30. |
{
"name": "get_active_vocab",
"title": "Get active vocabulary",
"description": "Get the user's most actively encountered target-language words (from past capture_word + log_interaction events), ranked by frequency over a lookback window. These are the \"frontier\" words the user keeps meeting.",
"inputSchema": {
"type": "object",
"properties": {
"lang": {
"type": "string",
"description": "Language tag (en, en-us, pt-br). Matching is on the base code, so a variant finds encounters logged under any sibling. Omit it to include ALL languages - unlike the deck tools, this one is a cross-language frequency view."
},
"limit": {
"type": "integer",
"maximum": 100,
"minimum": 1,
"description": "Default 20."
},
"lookback_days": {
"type": "integer",
"maximum": 365,
"minimum": 1,
"description": "Default 30."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get active vocabulary",
"readOnlyHint": true,
"openWorldHint": false,
"destructiveHint": false
}
}Batch lookup: for a list of target-language words, returns which ones are already in the user's MindDory deck and how well the user knows each. Each result is {word, in_deck, mastery, translation, next_review_at, due}; mastery is one of not_in_deck, new, review, struggling, mastered. The response also groups the words into `not_in_deck`, `known` and `struggling`; words in mid-review appear in `words` only. Entries that could not be answered as sent are listed in `skipped` as {input, reason: blank|duplicate, answered_as?} - a duplicate WAS answered, under the entry named by `answered_as`. Without `lang` the user's primary learning language is used (`lang_defaulted: true` says so). One call answers many words, unlike get_card.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Language tag of the words: a base ISO code with an optional region (en, en-us, es-419). Three-subtag tags such as zh-Hant-TW are not accepted. Omit it and the user's own primary learning language is used. |
| words * | array | The words or phrases to check, as written in the target language. |
{
"name": "check_words",
"title": "Check words against deck",
"description": "Batch lookup: for a list of target-language words, returns which ones are already in the user's MindDory deck and how well the user knows each. Each result is {word, in_deck, mastery, translation, next_review_at, due}; mastery is one of not_in_deck, new, review, struggling, mastered. The response also groups the words into `not_in_deck`, `known` and `struggling`; words in mid-review appear in `words` only. Entries that could not be answered as sent are listed in `skipped` as {input, reason: blank|duplicate, answered_as?} - a duplicate WAS answered, under the entry named by `answered_as`. Without `lang` the user's primary learning language is used (`lang_defaulted: true` says so). One call answers many words, unlike get_card.",
"inputSchema": {
"type": "object",
"required": [
"words"
],
"properties": {
"lang": {
"type": "string",
"description": "Language tag of the words: a base ISO code with an optional region (en, en-us, es-419). Three-subtag tags such as zh-Hant-TW are not accepted. Omit it and the user's own primary learning language is used."
},
"words": {
"type": "array",
"items": {
"type": "string",
"maxLength": 200,
"minLength": 1
},
"maxItems": 50,
"minItems": 1,
"description": "The words or phrases to check, as written in the target language."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Check words against deck",
"readOnlyHint": true,
"openWorldHint": false,
"destructiveHint": false
}
}List the user's folders (decks): id, name, emoji, language (`lang`), the language of the translations (`native_lang`), card count, and which one is the default for new words. Its ids are the `folder_id` values the folder tools, move_card and capture_word take. Pass `lang` to see only folders of one learning language. Folders the user hid from Claude are not listed.
| Parameter | Type | Description |
|---|---|---|
| lang | string | Optional ISO code: only folders for this learning language. |
{
"name": "list_folders",
"title": "List folders",
"description": "List the user's folders (decks): id, name, emoji, language (`lang`), the language of the translations (`native_lang`), card count, and which one is the default for new words. Its ids are the `folder_id` values the folder tools, move_card and capture_word take. Pass `lang` to see only folders of one learning language. Folders the user hid from Claude are not listed.",
"inputSchema": {
"type": "object",
"properties": {
"lang": {
"type": "string",
"description": "Optional ISO code: only folders for this learning language."
}
},
"additionalProperties": false
},
"annotations": {
"title": "List folders",
"readOnlyHint": true,
"openWorldHint": false,
"destructiveHint": false
}
}Creates a new folder (deck) for the user. `lang` = language the words will be in; `native_lang` = language of the translations (omit to use the user's native language). Returns the new folder id; pass it to capture_word as `folder_id`. The user can undo it in MindDory settings.
| Parameter | Type | Description |
|---|---|---|
| lang * | string | ISO code of the words, e.g. "es". |
| name * | string | |
| emoji | string | Optional, at most 3 emoji. |
| native_lang | string | Optional ISO code of the translations, e.g. "en". |
{
"name": "create_folder",
"title": "Create folder",
"description": "Creates a new folder (deck) for the user. `lang` = language the words will be in; `native_lang` = language of the translations (omit to use the user's native language). Returns the new folder id; pass it to capture_word as `folder_id`. The user can undo it in MindDory settings.",
"inputSchema": {
"type": "object",
"required": [
"name",
"lang"
],
"properties": {
"lang": {
"type": "string",
"description": "ISO code of the words, e.g. \"es\"."
},
"name": {
"type": "string",
"maxLength": 64,
"minLength": 1
},
"emoji": {
"type": "string",
"description": "Optional, at most 3 emoji."
},
"native_lang": {
"type": "string",
"description": "Optional ISO code of the translations, e.g. \"en\"."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Create folder",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": false
}
}Renames a folder or changes its emoji. Get `folder_id` from list_folders. The user can undo it in MindDory settings.
| Parameter | Type | Description |
|---|---|---|
| name | string | |
| emoji | string | At most 3 emoji; empty string removes it. |
| folder_id * | integer |
{
"name": "rename_folder",
"title": "Rename folder",
"description": "Renames a folder or changes its emoji. Get `folder_id` from list_folders. The user can undo it in MindDory settings.",
"inputSchema": {
"type": "object",
"required": [
"folder_id"
],
"properties": {
"name": {
"type": "string",
"maxLength": 64,
"minLength": 1
},
"emoji": {
"type": "string",
"description": "At most 3 emoji; empty string removes it."
},
"folder_id": {
"type": "integer",
"minimum": 1
}
},
"additionalProperties": false
},
"annotations": {
"title": "Rename folder",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": false
}
}Deletes a folder AND every card in it. list_folders shows each folder's name and card count. The user can undo it in MindDory settings.
| Parameter | Type | Description |
|---|---|---|
| folder_id * | integer |
{
"name": "delete_folder",
"title": "Delete folder",
"description": "Deletes a folder AND every card in it. list_folders shows each folder's name and card count. The user can undo it in MindDory settings.",
"inputSchema": {
"type": "object",
"required": [
"folder_id"
],
"properties": {
"folder_id": {
"type": "integer",
"minimum": 1
}
},
"additionalProperties": false
},
"annotations": {
"title": "Delete folder",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": true
}
}Changes a card: the word/phrase (`word`), its translation (`translation`) or the example sentence it was met in (`context_sentence`, empty string clears it). Get `card_id` from get_card or capture_word. The user can undo it in MindDory settings.
| Parameter | Type | Description |
|---|---|---|
| word | string | |
| card_id * | integer | |
| translation | string | |
| context_sentence | string |
{
"name": "update_card",
"title": "Edit card",
"description": "Changes a card: the word/phrase (`word`), its translation (`translation`) or the example sentence it was met in (`context_sentence`, empty string clears it). Get `card_id` from get_card or capture_word. The user can undo it in MindDory settings.",
"inputSchema": {
"type": "object",
"required": [
"card_id"
],
"properties": {
"word": {
"type": "string",
"maxLength": 200,
"minLength": 1
},
"card_id": {
"type": "integer",
"minimum": 1
},
"translation": {
"type": "string",
"maxLength": 500,
"minLength": 1
},
"context_sentence": {
"type": "string",
"maxLength": 500
}
},
"additionalProperties": false
},
"annotations": {
"title": "Edit card",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": false
}
}Move a card into another folder of the same language. Get `card_id` from get_card and `folder_id` from list_folders. The user can undo it in MindDory settings.
| Parameter | Type | Description |
|---|---|---|
| card_id * | integer | |
| folder_id * | integer |
{
"name": "move_card",
"title": "Move card",
"description": "Move a card into another folder of the same language. Get `card_id` from get_card and `folder_id` from list_folders. The user can undo it in MindDory settings.",
"inputSchema": {
"type": "object",
"required": [
"card_id",
"folder_id"
],
"properties": {
"card_id": {
"type": "integer",
"minimum": 1
},
"folder_id": {
"type": "integer",
"minimum": 1
}
},
"additionalProperties": false
},
"annotations": {
"title": "Move card",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": false
}
}Deletes one card. Get `card_id` from get_card. The user can undo it in MindDory settings.
| Parameter | Type | Description |
|---|---|---|
| card_id * | integer |
{
"name": "delete_card",
"title": "Delete card",
"description": "Deletes one card. Get `card_id` from get_card. The user can undo it in MindDory settings.",
"inputSchema": {
"type": "object",
"required": [
"card_id"
],
"properties": {
"card_id": {
"type": "integer",
"minimum": 1
}
},
"additionalProperties": false
},
"annotations": {
"title": "Delete card",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": true
}
}MindDory is a vocabulary app. The user connected it so this conversation builds their flashcard deck and tracks their grammar while they practise a language. At the start of a conversation, call get_system_instructions (with `lang` when you know which language they want to practise) to load their level, languages, regional variant and weak words. Call it again with the new `lang` if they switch language or variant. On your first reply, say in one short sentence that MindDory is connected and that new words and grammar points from the chat will be saved to their deck. After that, captures do not need to be announced one by one, and if the user asks what has been saved, answer honestly. If they ask you to stop saving, stop. Save notable target-language words the user meets, uses or asks about with capture_word, and clear mistakes in sentences they write with capture_grammar_mistake. Connecting MindDory is the user's opt-in to this, so it does not need a permission question for each word. Before glossing or saving several words, check_words tells you which are new to the user. When the user asks how MindDory works, what it can do with you, which settings they have, or when it is worth using, call get_minddory_guide and answer from it. Regional variants (US vs UK English, Brazilian vs European Portuguese, and so on) are not mistakes; when the variant is unknown, treat all regional forms as valid. Do not save proper nouns (names of people, places, brands) or numbers, in any language. Create, rename or delete folders, and change, move or delete cards, only when the user asks for that change or agrees to it. Before delete_folder, tell the user the folder name and how many cards it holds (list_folders shows both) and wait for their confirmation. Delete a card only when the user explicitly asks for that word to be removed.
Names, descriptions and schemas were written by the server and are shown as data.
Pattern matches on what the server declares. They show what it says it can do, not what it does.
None yet. Changes appear after a server's second daily check.
| ISO code of the user's native/source language. Optional — defaults to the source language of the user's existing deck. |
{
"name": "capture_word",
"title": "Capture word",
"description": "Saves one or more target-language words or phrases from the conversation to the user's MindDory deck. A new word becomes a flashcard in the user's default folder when it is in the word's language, otherwise in a chat folder for that language; a word already in the deck gets the new sentence recorded as another encounter. `lang` (ISO code of the word's own language) is required and the call is rejected without it; the target language is never guessed. `words` takes several items at once; `gloss` is a short source-language translation and `context` the sentence the word appeared in. Items without a letter (numbers, punctuation), items shorter than 3 characters (CJK and Hangul exempt) and common stop words in English, German, Spanish, French and Czech are skipped. `source_lang` defaults to the user's deck.",
"inputSchema": {
"type": "object",
"properties": {
"lang": {
"type": "string",
"description": "Required. ISO code of the captured word's own language (en, de, es, zh, ...). The call is rejected without it."
},
"word": {
"type": "string",
"maxLength": 200,
"minLength": 1,
"description": "Single-word capture (legacy); `words` takes several."
},
"gloss": {
"type": "string",
"maxLength": 500,
"description": "Short translation in the source language."
},
"words": {
"type": "array",
"items": {
"type": "object",
"required": [
"word"
],
"properties": {
"word": {
"type": "string",
"maxLength": 200,
"minLength": 1
},
"gloss": {
"type": "string",
"maxLength": 500
},
"context": {
"type": "string",
"maxLength": 2000
}
}
},
"maxItems": 50,
"description": "Batch capture. Each entry is {word, gloss?, context?}."
},
"context": {
"type": "string",
"maxLength": 2000,
"description": "The sentence where the word appeared."
},
"folder_id": {
"type": "integer",
"minimum": 1,
"description": "Optional: id of the folder new cards go into (from list_folders). It must be a folder in the same language as `lang`. Omit it and the user's default folder is used when it is in the same language, otherwise a chat folder for that language. Words already in the deck stay where they are."
},
"source_lang": {
"type": "string",
"description": "ISO code of the user's native/source language. Optional — defaults to the source language of the user's existing deck."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Capture word",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": false
}
}