@mastra/mongodb 0.0.2-alpha.2 → 0.0.2-alpha.4

package/.turbo/turbo-build.log

@@ -1,23 +1,23 @@
 
-> @mastra/mongodb@0.0.2-alpha.2 build /home/runner/work/mastra/mastra/stores/mongodb
+> @mastra/mongodb@0.0.2-alpha.4 build /home/runner/work/mastra/mastra/stores/mongodb
 > tsup src/index.ts --format esm,cjs --experimental-dts --clean --treeshake=smallest --splitting
 
 CLI Building entry: src/index.ts
 CLI Using tsconfig: tsconfig.json
 CLI tsup v8.4.0
 TSC Build start
-TSC ⚡️ Build success in 8226ms
+TSC ⚡️ Build success in 7155ms
 DTS Build start
 CLI Target: es2022
-Analysis will use the bundled TypeScript version 5.8.2
+Analysis will use the bundled TypeScript version 5.8.3
 Writing package typings: /home/runner/work/mastra/mastra/stores/mongodb/dist/_tsup-dts-rollup.d.ts
-Analysis will use the bundled TypeScript version 5.8.2
+Analysis will use the bundled TypeScript version 5.8.3
 Writing package typings: /home/runner/work/mastra/mastra/stores/mongodb/dist/_tsup-dts-rollup.d.cts
-DTS ⚡️ Build success in 9773ms
+DTS ⚡️ Build success in 9430ms
 CLI Cleaning output folder
 ESM Build start
 CJS Build start
-CJS dist/index.cjs 12.33 KB
-CJS ⚡️ Build success in 633ms
-ESM dist/index.js 12.29 KB
-ESM ⚡️ Build success in 634ms
+ESM dist/index.js 16.22 KB
+ESM ⚡️ Build success in 751ms
+CJS dist/index.cjs 16.28 KB
+CJS ⚡️ Build success in 751ms

package/CHANGELOG.md

@@ -1,5 +1,21 @@
 # @mastra/mongodb
 
+## 0.0.2-alpha.4
+
+### Patch Changes
+
+- 5f826d9: Moved vector store specific prompts from @mastra/rag to be exported from the store that the prompt belongs to, ie @mastra/pg
+- Updated dependencies [3e7b69d]
+  - @mastra/core@0.9.1-alpha.5
+
+## 0.0.2-alpha.3
+
+### Patch Changes
+
+- Updated dependencies [e4943b8]
+- Updated dependencies [479f490]
+  - @mastra/core@0.9.1-alpha.4
+
 ## 0.0.2-alpha.2
 
 ### Patch Changes

package/dist/_tsup-dts-rollup.d.cts

@@ -12,6 +12,14 @@ import type { UpsertVectorArgs } from '@mastra/core/vector';
 import type { UpsertVectorParams } from '@mastra/core/vector';
 import type { VectorFilter } from '@mastra/core/vector/filter';
 
+/**
+ * Vector store specific prompt that details supported operators and examples.
+ * This prompt helps users construct valid filters for MongoDB Vector.
+ */
+declare const MONGODB_PROMPT = "When querying MongoDB Vector, you can ONLY use the operators listed below. Any other operators will be rejected.\nImportant: Don't explain how to construct the filter - use the specified operators and fields to search the content and return relevant results.\nIf a user tries to give an explicit operator that is not supported, reject the filter entirely and let them know that the operator is not supported.\n\nBasic Comparison Operators:\n- $eq: Exact match (default when using field: value)\n  Example: { \"category\": \"electronics\" }\n- $ne: Not equal\n  Example: { \"category\": { \"$ne\": \"electronics\" } }\n- $gt: Greater than\n  Example: { \"price\": { \"$gt\": 100 } }\n- $gte: Greater than or equal\n  Example: { \"price\": { \"$gte\": 100 } }\n- $lt: Less than\n  Example: { \"price\": { \"$lt\": 100 } }\n- $lte: Less than or equal\n  Example: { \"price\": { \"$lte\": 100 } }\n\nArray Operators:\n- $in: Match any value in array\n  Example: { \"category\": { \"$in\": [\"electronics\", \"books\"] } }\n- $nin: Does not match any value in array\n  Example: { \"category\": { \"$nin\": [\"electronics\", \"books\"] } }\n- $all: Match all values in array\n  Example: { \"tags\": { \"$all\": [\"premium\", \"sale\"] } }\n- $elemMatch: Match array elements that meet all specified conditions\n  Example: { \"items\": { \"$elemMatch\": { \"price\": { \"$gt\": 100 } } } }\n- $size: Match arrays with specific length\n  Example: { \"tags\": { \"$size\": 3 } }\n\nLogical Operators:\n- $and: Logical AND (can be implicit or explicit)\n  Implicit Example: { \"price\": { \"$gt\": 100 }, \"category\": \"electronics\" }\n  Explicit Example: { \"$and\": [{ \"price\": { \"$gt\": 100 } }, { \"category\": \"electronics\" }] }\n- $or: Logical OR\n  Example: { \"$or\": [{ \"price\": { \"$lt\": 50 } }, { \"category\": \"books\" }] }\n- $not: Logical NOT\n  Example: { \"$not\": { \"category\": \"electronics\" } }\n- $nor: Logical NOR\n  Example: { \"$nor\": [{ \"price\": { \"$lt\": 50 } }, { \"category\": \"books\" }] }\n\nElement Operators:\n- $exists: Check if field exists\n  Example: { \"rating\": { \"$exists\": true } }\n- $type: Check field type\n  Example: { \"price\": { \"$type\": \"number\" } }\n\nText Search Operators:\n- $text: Full text search\n  Example: { \"$text\": { \"$search\": \"gaming laptop\" } }\n- $regex: Regular expression match\n  Example: { \"name\": { \"$regex\": \"^Gaming\" } }\n\nRestrictions:\n- Only logical operators ($and, $or, $not, $nor) can be used at the top level\n- Empty arrays in array operators will return no results\n- Nested fields are supported using dot notation\n- Multiple conditions on the same field are supported\n- At least one key-value pair is required in filter object\n- Empty objects and undefined values are treated as no filter\n- Invalid types in comparison operators will throw errors\n- All non-logical operators must be used within a field condition\n  Valid: { \"field\": { \"$gt\": 100 } }\n  Valid: { \"$and\": [...] }\n  Invalid: { \"$gt\": 100 }\n- Logical operators must contain field conditions, not direct operators\n  Valid: { \"$and\": [{ \"field\": { \"$gt\": 100 } }] }\n  Invalid: { \"$and\": [{ \"$gt\": 100 }] }\n- Logical operators ($and, $or, $not, $nor):\n  - Can only be used at top level or nested within other logical operators\n  - Can not be used on a field level, or be nested inside a field\n  - Can not be used inside an operator\n  - Valid: { \"$and\": [{ \"field\": { \"$gt\": 100 } }] }\n  - Valid: { \"$or\": [{ \"$and\": [{ \"field\": { \"$gt\": 100 } }] }] }\n  - Invalid: { \"field\": { \"$and\": [{ \"$gt\": 100 }] } }\n  - Invalid: { \"field\": { \"$or\": [{ \"$gt\": 100 }] } }\n  - Invalid: { \"field\": { \"$gt\": { \"$and\": [{...}] } } }\n\nExample Complex Query:\n{\n  \"$and\": [\n    { \"category\": { \"$in\": [\"electronics\", \"computers\"] } },\n    { \"price\": { \"$gte\": 100, \"$lte\": 1000 } },\n    { \"tags\": { \"$all\": [\"premium\", \"sale\"] } },\n    { \"items\": { \"$elemMatch\": { \"price\": { \"$gt\": 50 }, \"inStock\": true } } },\n    { \"$text\": { \"$search\": \"gaming laptop\" } },\n    { \"$or\": [\n      { \"stock\": { \"$gt\": 0 } },\n      { \"preorder\": true }\n    ]},\n    { \"$not\": { \"status\": \"discontinued\" } }\n  ]\n}";
+export { MONGODB_PROMPT }
+export { MONGODB_PROMPT as MONGODB_PROMPT_alias_1 }
+
 /**
  * Translator for MongoDB filter queries.
  * Maintains MongoDB-compatible syntax while ensuring proper validation

package/dist/_tsup-dts-rollup.d.ts

@@ -12,6 +12,14 @@ import type { UpsertVectorArgs } from '@mastra/core/vector';
 import type { UpsertVectorParams } from '@mastra/core/vector';
 import type { VectorFilter } from '@mastra/core/vector/filter';
 
+/**
+ * Vector store specific prompt that details supported operators and examples.
+ * This prompt helps users construct valid filters for MongoDB Vector.
+ */
+declare const MONGODB_PROMPT = "When querying MongoDB Vector, you can ONLY use the operators listed below. Any other operators will be rejected.\nImportant: Don't explain how to construct the filter - use the specified operators and fields to search the content and return relevant results.\nIf a user tries to give an explicit operator that is not supported, reject the filter entirely and let them know that the operator is not supported.\n\nBasic Comparison Operators:\n- $eq: Exact match (default when using field: value)\n  Example: { \"category\": \"electronics\" }\n- $ne: Not equal\n  Example: { \"category\": { \"$ne\": \"electronics\" } }\n- $gt: Greater than\n  Example: { \"price\": { \"$gt\": 100 } }\n- $gte: Greater than or equal\n  Example: { \"price\": { \"$gte\": 100 } }\n- $lt: Less than\n  Example: { \"price\": { \"$lt\": 100 } }\n- $lte: Less than or equal\n  Example: { \"price\": { \"$lte\": 100 } }\n\nArray Operators:\n- $in: Match any value in array\n  Example: { \"category\": { \"$in\": [\"electronics\", \"books\"] } }\n- $nin: Does not match any value in array\n  Example: { \"category\": { \"$nin\": [\"electronics\", \"books\"] } }\n- $all: Match all values in array\n  Example: { \"tags\": { \"$all\": [\"premium\", \"sale\"] } }\n- $elemMatch: Match array elements that meet all specified conditions\n  Example: { \"items\": { \"$elemMatch\": { \"price\": { \"$gt\": 100 } } } }\n- $size: Match arrays with specific length\n  Example: { \"tags\": { \"$size\": 3 } }\n\nLogical Operators:\n- $and: Logical AND (can be implicit or explicit)\n  Implicit Example: { \"price\": { \"$gt\": 100 }, \"category\": \"electronics\" }\n  Explicit Example: { \"$and\": [{ \"price\": { \"$gt\": 100 } }, { \"category\": \"electronics\" }] }\n- $or: Logical OR\n  Example: { \"$or\": [{ \"price\": { \"$lt\": 50 } }, { \"category\": \"books\" }] }\n- $not: Logical NOT\n  Example: { \"$not\": { \"category\": \"electronics\" } }\n- $nor: Logical NOR\n  Example: { \"$nor\": [{ \"price\": { \"$lt\": 50 } }, { \"category\": \"books\" }] }\n\nElement Operators:\n- $exists: Check if field exists\n  Example: { \"rating\": { \"$exists\": true } }\n- $type: Check field type\n  Example: { \"price\": { \"$type\": \"number\" } }\n\nText Search Operators:\n- $text: Full text search\n  Example: { \"$text\": { \"$search\": \"gaming laptop\" } }\n- $regex: Regular expression match\n  Example: { \"name\": { \"$regex\": \"^Gaming\" } }\n\nRestrictions:\n- Only logical operators ($and, $or, $not, $nor) can be used at the top level\n- Empty arrays in array operators will return no results\n- Nested fields are supported using dot notation\n- Multiple conditions on the same field are supported\n- At least one key-value pair is required in filter object\n- Empty objects and undefined values are treated as no filter\n- Invalid types in comparison operators will throw errors\n- All non-logical operators must be used within a field condition\n  Valid: { \"field\": { \"$gt\": 100 } }\n  Valid: { \"$and\": [...] }\n  Invalid: { \"$gt\": 100 }\n- Logical operators must contain field conditions, not direct operators\n  Valid: { \"$and\": [{ \"field\": { \"$gt\": 100 } }] }\n  Invalid: { \"$and\": [{ \"$gt\": 100 }] }\n- Logical operators ($and, $or, $not, $nor):\n  - Can only be used at top level or nested within other logical operators\n  - Can not be used on a field level, or be nested inside a field\n  - Can not be used inside an operator\n  - Valid: { \"$and\": [{ \"field\": { \"$gt\": 100 } }] }\n  - Valid: { \"$or\": [{ \"$and\": [{ \"field\": { \"$gt\": 100 } }] }] }\n  - Invalid: { \"field\": { \"$and\": [{ \"$gt\": 100 }] } }\n  - Invalid: { \"field\": { \"$or\": [{ \"$gt\": 100 }] } }\n  - Invalid: { \"field\": { \"$gt\": { \"$and\": [{...}] } } }\n\nExample Complex Query:\n{\n  \"$and\": [\n    { \"category\": { \"$in\": [\"electronics\", \"computers\"] } },\n    { \"price\": { \"$gte\": 100, \"$lte\": 1000 } },\n    { \"tags\": { \"$all\": [\"premium\", \"sale\"] } },\n    { \"items\": { \"$elemMatch\": { \"price\": { \"$gt\": 50 }, \"inStock\": true } } },\n    { \"$text\": { \"$search\": \"gaming laptop\" } },\n    { \"$or\": [\n      { \"stock\": { \"$gt\": 0 } },\n      { \"preorder\": true }\n    ]},\n    { \"$not\": { \"status\": \"discontinued\" } }\n  ]\n}";
+export { MONGODB_PROMPT }
+export { MONGODB_PROMPT as MONGODB_PROMPT_alias_1 }
+
 /**
  * Translator for MongoDB filter queries.
  * Maintains MongoDB-compatible syntax while ensuring proper validation

package/dist/index.cjs

@@ -358,4 +358,100 @@ var MongoDBVector = class extends vector.MastraVector {
   }
 };
 
+// src/vector/prompt.ts
+var MONGODB_PROMPT = `When querying MongoDB Vector, you can ONLY use the operators listed below. Any other operators will be rejected.
+Important: Don't explain how to construct the filter - use the specified operators and fields to search the content and return relevant results.
+If a user tries to give an explicit operator that is not supported, reject the filter entirely and let them know that the operator is not supported.
+
+Basic Comparison Operators:
+- $eq: Exact match (default when using field: value)
+  Example: { "category": "electronics" }
+- $ne: Not equal
+  Example: { "category": { "$ne": "electronics" } }
+- $gt: Greater than
+  Example: { "price": { "$gt": 100 } }
+- $gte: Greater than or equal
+  Example: { "price": { "$gte": 100 } }
+- $lt: Less than
+  Example: { "price": { "$lt": 100 } }
+- $lte: Less than or equal
+  Example: { "price": { "$lte": 100 } }
+
+Array Operators:
+- $in: Match any value in array
+  Example: { "category": { "$in": ["electronics", "books"] } }
+- $nin: Does not match any value in array
+  Example: { "category": { "$nin": ["electronics", "books"] } }
+- $all: Match all values in array
+  Example: { "tags": { "$all": ["premium", "sale"] } }
+- $elemMatch: Match array elements that meet all specified conditions
+  Example: { "items": { "$elemMatch": { "price": { "$gt": 100 } } } }
+- $size: Match arrays with specific length
+  Example: { "tags": { "$size": 3 } }
+
+Logical Operators:
+- $and: Logical AND (can be implicit or explicit)
+  Implicit Example: { "price": { "$gt": 100 }, "category": "electronics" }
+  Explicit Example: { "$and": [{ "price": { "$gt": 100 } }, { "category": "electronics" }] }
+- $or: Logical OR
+  Example: { "$or": [{ "price": { "$lt": 50 } }, { "category": "books" }] }
+- $not: Logical NOT
+  Example: { "$not": { "category": "electronics" } }
+- $nor: Logical NOR
+  Example: { "$nor": [{ "price": { "$lt": 50 } }, { "category": "books" }] }
+
+Element Operators:
+- $exists: Check if field exists
+  Example: { "rating": { "$exists": true } }
+- $type: Check field type
+  Example: { "price": { "$type": "number" } }
+
+Text Search Operators:
+- $text: Full text search
+  Example: { "$text": { "$search": "gaming laptop" } }
+- $regex: Regular expression match
+  Example: { "name": { "$regex": "^Gaming" } }
+
+Restrictions:
+- Only logical operators ($and, $or, $not, $nor) can be used at the top level
+- Empty arrays in array operators will return no results
+- Nested fields are supported using dot notation
+- Multiple conditions on the same field are supported
+- At least one key-value pair is required in filter object
+- Empty objects and undefined values are treated as no filter
+- Invalid types in comparison operators will throw errors
+- All non-logical operators must be used within a field condition
+  Valid: { "field": { "$gt": 100 } }
+  Valid: { "$and": [...] }
+  Invalid: { "$gt": 100 }
+- Logical operators must contain field conditions, not direct operators
+  Valid: { "$and": [{ "field": { "$gt": 100 } }] }
+  Invalid: { "$and": [{ "$gt": 100 }] }
+- Logical operators ($and, $or, $not, $nor):
+  - Can only be used at top level or nested within other logical operators
+  - Can not be used on a field level, or be nested inside a field
+  - Can not be used inside an operator
+  - Valid: { "$and": [{ "field": { "$gt": 100 } }] }
+  - Valid: { "$or": [{ "$and": [{ "field": { "$gt": 100 } }] }] }
+  - Invalid: { "field": { "$and": [{ "$gt": 100 }] } }
+  - Invalid: { "field": { "$or": [{ "$gt": 100 }] } }
+  - Invalid: { "field": { "$gt": { "$and": [{...}] } } }
+
+Example Complex Query:
+{
+  "$and": [
+    { "category": { "$in": ["electronics", "computers"] } },
+    { "price": { "$gte": 100, "$lte": 1000 } },
+    { "tags": { "$all": ["premium", "sale"] } },
+    { "items": { "$elemMatch": { "price": { "$gt": 50 }, "inStock": true } } },
+    { "$text": { "$search": "gaming laptop" } },
+    { "$or": [
+      { "stock": { "$gt": 0 } },
+      { "preorder": true }
+    ]},
+    { "$not": { "status": "discontinued" } }
+  ]
+}`;
+
+exports.MONGODB_PROMPT = MONGODB_PROMPT;
 exports.MongoDBVector = MongoDBVector;

package/dist/index.d.cts

@@ -1,3 +1,4 @@
+export { MONGODB_PROMPT } from './_tsup-dts-rollup.cjs';
 export { MongoDBUpsertArgs } from './_tsup-dts-rollup.cjs';
 export { MongoDBQueryArgs } from './_tsup-dts-rollup.cjs';
 export { MongoDBUpsertParams } from './_tsup-dts-rollup.cjs';

package/dist/index.d.ts

@@ -1,3 +1,4 @@
+export { MONGODB_PROMPT } from './_tsup-dts-rollup.js';
 export { MongoDBUpsertArgs } from './_tsup-dts-rollup.js';
 export { MongoDBQueryArgs } from './_tsup-dts-rollup.js';
 export { MongoDBUpsertParams } from './_tsup-dts-rollup.js';

package/dist/index.js

@@ -356,4 +356,99 @@ var MongoDBVector = class extends MastraVector {
   }
 };
 
-export { MongoDBVector };
+// src/vector/prompt.ts
+var MONGODB_PROMPT = `When querying MongoDB Vector, you can ONLY use the operators listed below. Any other operators will be rejected.
+Important: Don't explain how to construct the filter - use the specified operators and fields to search the content and return relevant results.
+If a user tries to give an explicit operator that is not supported, reject the filter entirely and let them know that the operator is not supported.
+
+Basic Comparison Operators:
+- $eq: Exact match (default when using field: value)
+  Example: { "category": "electronics" }
+- $ne: Not equal
+  Example: { "category": { "$ne": "electronics" } }
+- $gt: Greater than
+  Example: { "price": { "$gt": 100 } }
+- $gte: Greater than or equal
+  Example: { "price": { "$gte": 100 } }
+- $lt: Less than
+  Example: { "price": { "$lt": 100 } }
+- $lte: Less than or equal
+  Example: { "price": { "$lte": 100 } }
+
+Array Operators:
+- $in: Match any value in array
+  Example: { "category": { "$in": ["electronics", "books"] } }
+- $nin: Does not match any value in array
+  Example: { "category": { "$nin": ["electronics", "books"] } }
+- $all: Match all values in array
+  Example: { "tags": { "$all": ["premium", "sale"] } }
+- $elemMatch: Match array elements that meet all specified conditions
+  Example: { "items": { "$elemMatch": { "price": { "$gt": 100 } } } }
+- $size: Match arrays with specific length
+  Example: { "tags": { "$size": 3 } }
+
+Logical Operators:
+- $and: Logical AND (can be implicit or explicit)
+  Implicit Example: { "price": { "$gt": 100 }, "category": "electronics" }
+  Explicit Example: { "$and": [{ "price": { "$gt": 100 } }, { "category": "electronics" }] }
+- $or: Logical OR
+  Example: { "$or": [{ "price": { "$lt": 50 } }, { "category": "books" }] }
+- $not: Logical NOT
+  Example: { "$not": { "category": "electronics" } }
+- $nor: Logical NOR
+  Example: { "$nor": [{ "price": { "$lt": 50 } }, { "category": "books" }] }
+
+Element Operators:
+- $exists: Check if field exists
+  Example: { "rating": { "$exists": true } }
+- $type: Check field type
+  Example: { "price": { "$type": "number" } }
+
+Text Search Operators:
+- $text: Full text search
+  Example: { "$text": { "$search": "gaming laptop" } }
+- $regex: Regular expression match
+  Example: { "name": { "$regex": "^Gaming" } }
+
+Restrictions:
+- Only logical operators ($and, $or, $not, $nor) can be used at the top level
+- Empty arrays in array operators will return no results
+- Nested fields are supported using dot notation
+- Multiple conditions on the same field are supported
+- At least one key-value pair is required in filter object
+- Empty objects and undefined values are treated as no filter
+- Invalid types in comparison operators will throw errors
+- All non-logical operators must be used within a field condition
+  Valid: { "field": { "$gt": 100 } }
+  Valid: { "$and": [...] }
+  Invalid: { "$gt": 100 }
+- Logical operators must contain field conditions, not direct operators
+  Valid: { "$and": [{ "field": { "$gt": 100 } }] }
+  Invalid: { "$and": [{ "$gt": 100 }] }
+- Logical operators ($and, $or, $not, $nor):
+  - Can only be used at top level or nested within other logical operators
+  - Can not be used on a field level, or be nested inside a field
+  - Can not be used inside an operator
+  - Valid: { "$and": [{ "field": { "$gt": 100 } }] }
+  - Valid: { "$or": [{ "$and": [{ "field": { "$gt": 100 } }] }] }
+  - Invalid: { "field": { "$and": [{ "$gt": 100 }] } }
+  - Invalid: { "field": { "$or": [{ "$gt": 100 }] } }
+  - Invalid: { "field": { "$gt": { "$and": [{...}] } } }
+
+Example Complex Query:
+{
+  "$and": [
+    { "category": { "$in": ["electronics", "computers"] } },
+    { "price": { "$gte": 100, "$lte": 1000 } },
+    { "tags": { "$all": ["premium", "sale"] } },
+    { "items": { "$elemMatch": { "price": { "$gt": 50 }, "inStock": true } } },
+    { "$text": { "$search": "gaming laptop" } },
+    { "$or": [
+      { "stock": { "$gt": 0 } },
+      { "preorder": true }
+    ]},
+    { "$not": { "status": "discontinued" } }
+  ]
+}`;
+
+export { MONGODB_PROMPT, MongoDBVector };

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@mastra/mongodb",
-  "version": "0.0.2-alpha.2",
+  "version": "0.0.2-alpha.4",
   "description": "MongoDB provider for Mastra - includes vector store capabilities",
   "type": "module",
   "main": "dist/index.js",
@@ -23,15 +23,15 @@
     "cloudflare": "^4.1.0",
     "mongodb": "^6.15.0",
     "uuid": "^11.1.0",
-    "@mastra/core": "^0.9.1-alpha.3"
+    "@mastra/core": "^0.9.1-alpha.5"
   },
   "devDependencies": {
-    "@microsoft/api-extractor": "^7.52.1",
+    "@microsoft/api-extractor": "^7.52.5",
     "@types/node": "^20.17.27",
     "eslint": "^9.23.0",
     "tsup": "^8.4.0",
     "typescript": "^5.8.2",
-    "vitest": "^3.0.9",
+    "vitest": "^3.1.2",
     "@internal/lint": "0.0.2"
   },
   "scripts": {

package/src/index.ts

@@ -1 +1,2 @@
 export * from './vector';
+export { MONGODB_PROMPT } from './vector/prompt';

package/src/vector/prompt.ts

@@ -0,0 +1,97 @@
+/**
+ * Vector store specific prompt that details supported operators and examples.
+ * This prompt helps users construct valid filters for MongoDB Vector.
+ */
+export const MONGODB_PROMPT = `When querying MongoDB Vector, you can ONLY use the operators listed below. Any other operators will be rejected.
+Important: Don't explain how to construct the filter - use the specified operators and fields to search the content and return relevant results.
+If a user tries to give an explicit operator that is not supported, reject the filter entirely and let them know that the operator is not supported.
+
+Basic Comparison Operators:
+- $eq: Exact match (default when using field: value)
+  Example: { "category": "electronics" }
+- $ne: Not equal
+  Example: { "category": { "$ne": "electronics" } }
+- $gt: Greater than
+  Example: { "price": { "$gt": 100 } }
+- $gte: Greater than or equal
+  Example: { "price": { "$gte": 100 } }
+- $lt: Less than
+  Example: { "price": { "$lt": 100 } }
+- $lte: Less than or equal
+  Example: { "price": { "$lte": 100 } }
+
+Array Operators:
+- $in: Match any value in array
+  Example: { "category": { "$in": ["electronics", "books"] } }
+- $nin: Does not match any value in array
+  Example: { "category": { "$nin": ["electronics", "books"] } }
+- $all: Match all values in array
+  Example: { "tags": { "$all": ["premium", "sale"] } }
+- $elemMatch: Match array elements that meet all specified conditions
+  Example: { "items": { "$elemMatch": { "price": { "$gt": 100 } } } }
+- $size: Match arrays with specific length
+  Example: { "tags": { "$size": 3 } }
+
+Logical Operators:
+- $and: Logical AND (can be implicit or explicit)
+  Implicit Example: { "price": { "$gt": 100 }, "category": "electronics" }
+  Explicit Example: { "$and": [{ "price": { "$gt": 100 } }, { "category": "electronics" }] }
+- $or: Logical OR
+  Example: { "$or": [{ "price": { "$lt": 50 } }, { "category": "books" }] }
+- $not: Logical NOT
+  Example: { "$not": { "category": "electronics" } }
+- $nor: Logical NOR
+  Example: { "$nor": [{ "price": { "$lt": 50 } }, { "category": "books" }] }
+
+Element Operators:
+- $exists: Check if field exists
+  Example: { "rating": { "$exists": true } }
+- $type: Check field type
+  Example: { "price": { "$type": "number" } }
+
+Text Search Operators:
+- $text: Full text search
+  Example: { "$text": { "$search": "gaming laptop" } }
+- $regex: Regular expression match
+  Example: { "name": { "$regex": "^Gaming" } }
+
+Restrictions:
+- Only logical operators ($and, $or, $not, $nor) can be used at the top level
+- Empty arrays in array operators will return no results
+- Nested fields are supported using dot notation
+- Multiple conditions on the same field are supported
+- At least one key-value pair is required in filter object
+- Empty objects and undefined values are treated as no filter
+- Invalid types in comparison operators will throw errors
+- All non-logical operators must be used within a field condition
+  Valid: { "field": { "$gt": 100 } }
+  Valid: { "$and": [...] }
+  Invalid: { "$gt": 100 }
+- Logical operators must contain field conditions, not direct operators
+  Valid: { "$and": [{ "field": { "$gt": 100 } }] }
+  Invalid: { "$and": [{ "$gt": 100 }] }
+- Logical operators ($and, $or, $not, $nor):
+  - Can only be used at top level or nested within other logical operators
+  - Can not be used on a field level, or be nested inside a field
+  - Can not be used inside an operator
+  - Valid: { "$and": [{ "field": { "$gt": 100 } }] }
+  - Valid: { "$or": [{ "$and": [{ "field": { "$gt": 100 } }] }] }
+  - Invalid: { "field": { "$and": [{ "$gt": 100 }] } }
+  - Invalid: { "field": { "$or": [{ "$gt": 100 }] } }
+  - Invalid: { "field": { "$gt": { "$and": [{...}] } } }
+
+Example Complex Query:
+{
+  "$and": [
+    { "category": { "$in": ["electronics", "computers"] } },
+    { "price": { "$gte": 100, "$lte": 1000 } },
+    { "tags": { "$all": ["premium", "sale"] } },
+    { "items": { "$elemMatch": { "price": { "$gt": 50 }, "inStock": true } } },
+    { "$text": { "$search": "gaming laptop" } },
+    { "$or": [
+      { "stock": { "$gt": 0 } },
+      { "preorder": true }
+    ]},
+    { "$not": { "status": "discontinued" } }
+  ]
+}`;