@mindstudio-ai/agent 0.1.87 → 0.1.88

package/dist/index.d.ts

@@ -6869,7 +6869,7 @@ interface StepMethods {
      */
     convertPdfToImages(step: ConvertPdfToImagesStepInput, options?: StepExecutionOptions): Promise<StepExecutionResult<ConvertPdfToImagesStepOutput>>;
     /**
-     * Create a new empty vector data source for the current app.
+     * @deprecated Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Create a new empty vector data source for the current app.
      *
      * @remarks
      * - Creates a new data source (vector database) associated with the current app version.
@@ -6951,7 +6951,7 @@ interface StepMethods {
      */
     createGoogleSheet(step: CreateGoogleSheetStepInput, options?: StepExecutionOptions): Promise<StepExecutionResult<CreateGoogleSheetStepOutput>>;
     /**
-     * Delete a vector data source from the current app.
+     * @deprecated Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Delete a vector data source from the current app.
      *
      * @remarks
      * - Soft-deletes a data source (vector database) by marking it as deleted.
@@ -6967,7 +6967,7 @@ interface StepMethods {
      */
     deleteDataSource(step: DeleteDataSourceStepInput, options?: StepExecutionOptions): Promise<StepExecutionResult<DeleteDataSourceStepOutput>>;
     /**
-     * Delete a single document from a data source.
+     * @deprecated Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Delete a single document from a data source.
      *
      * @remarks
      * - Soft-deletes a document by marking it as deleted.
@@ -7226,7 +7226,7 @@ interface StepMethods {
      */
     extractText(step: ExtractTextStepInput, options?: StepExecutionOptions): Promise<StepExecutionResult<ExtractTextStepOutput>>;
     /**
-     * Fetch the full extracted text contents of a document in a data source.
+     * @deprecated Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Fetch the full extracted text contents of a document in a data source.
      *
      * @remarks
      * - Loads a document by ID and returns its full extracted text content.
@@ -7800,7 +7800,7 @@ interface StepMethods {
      */
     insertVideoClips(step: InsertVideoClipsStepInput, options?: StepExecutionOptions): Promise<StepExecutionResult<InsertVideoClipsStepOutput>>;
     /**
-     * List all data sources for the current app.
+     * @deprecated Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. List all data sources for the current app.
      *
      * @remarks
      * - Returns metadata for every data source associated with the current app version.
@@ -8343,7 +8343,7 @@ interface StepMethods {
      */
     queryAppDatabase(step: QueryAppDatabaseStepInput, options?: StepExecutionOptions): Promise<StepExecutionResult<QueryAppDatabaseStepOutput>>;
     /**
-     * Search a vector data source (RAG) and return relevant document chunks.
+     * @deprecated Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Search a vector data source (RAG) and return relevant document chunks.
      *
      * @remarks
      * - Queries a vectorized data source and returns the most relevant chunks.
@@ -9173,7 +9173,7 @@ interface StepMethods {
      */
     updateGoogleSheet(step: UpdateGoogleSheetStepInput, options?: StepExecutionOptions): Promise<StepExecutionResult<UpdateGoogleSheetStepOutput>>;
     /**
-     * Upload a file into an existing data source from a URL or raw text content.
+     * @deprecated Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Upload a file into an existing data source from a URL or raw text content.
      *
      * @remarks
      * - If "file" is a single URL, the file is downloaded from that URL and uploaded.

package/dist/index.js

@@ -2362,7 +2362,7 @@ var stepMetadata = {
   },
   "createDataSource": {
     stepType: "createDataSource",
-    description: "Create a new empty vector data source for the current app.",
+    description: "Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Create a new empty vector data source for the current app.",
     usageNotes: '- Creates a new data source (vector database) associated with the current app version.\n- The data source is created empty \u2014 use the "Upload Data Source Document" block to add documents.\n- Returns the new data source ID which can be used in subsequent blocks.',
     inputSchema: { "type": "object", "properties": { "name": { "type": "string", "description": "Name for the new data source (supports variable interpolation)" } }, "required": ["name"] },
     outputSchema: { "description": "This step does not produce output data." }
@@ -2399,14 +2399,14 @@ var stepMetadata = {
   },
   "deleteDataSource": {
     stepType: "deleteDataSource",
-    description: "Delete a vector data source from the current app.",
+    description: "Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Delete a vector data source from the current app.",
     usageNotes: "- Soft-deletes a data source (vector database) by marking it as deleted.\n- The Milvus partition is cleaned up asynchronously by a background cron job.\n- The data source must belong to the current app version.",
     inputSchema: { "type": "object", "properties": { "dataSourceId": { "type": "string", "description": "ID of the data source to delete (supports variable interpolation)" } }, "required": ["dataSourceId"] },
     outputSchema: { "description": "This step does not produce output data." }
   },
   "deleteDataSourceDocument": {
     stepType: "deleteDataSourceDocument",
-    description: "Delete a single document from a data source.",
+    description: "Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Delete a single document from a data source.",
     usageNotes: "- Soft-deletes a document by marking it as deleted.\n- Requires both the data source ID and document ID.\n- After deletion, reloads vectors into Milvus so the data source reflects the change immediately.",
     inputSchema: { "type": "object", "properties": { "dataSourceId": { "type": "string", "description": "ID of the data source containing the document (supports variable interpolation)" }, "documentId": { "type": "string", "description": "ID of the document to delete (supports variable interpolation)" } }, "required": ["dataSourceId", "documentId"] },
     outputSchema: { "description": "This step does not produce output data." }
@@ -2511,7 +2511,7 @@ var stepMetadata = {
   },
   "fetchDataSourceDocument": {
     stepType: "fetchDataSourceDocument",
-    description: "Fetch the full extracted text contents of a document in a data source.",
+    description: "Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Fetch the full extracted text contents of a document in a data source.",
     usageNotes: '- Loads a document by ID and returns its full extracted text content.\n- The document must have been successfully processed (status "done").\n- Also returns document metadata (name, summary, word count).',
     inputSchema: { "type": "object", "properties": { "dataSourceId": { "type": "string", "description": "ID of the data source containing the document (supports variable interpolation)" }, "documentId": { "type": "string", "description": "ID of the document to fetch (supports variable interpolation)" } }, "required": ["dataSourceId", "documentId"] },
     outputSchema: { "description": "This step does not produce output data." }
@@ -2788,7 +2788,7 @@ var stepMetadata = {
   },
   "listDataSources": {
     stepType: "listDataSources",
-    description: "List all data sources for the current app.",
+    description: "Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. List all data sources for the current app.",
     usageNotes: "- Returns metadata for every data source associated with the current app version.\n- Each entry includes the data source ID, name, description, status, and document list.",
     inputSchema: { "type": "object", "properties": {}, "required": [] },
     outputSchema: { "description": "This step does not produce output data." }
@@ -3023,7 +3023,7 @@ var stepMetadata = {
   },
   "queryDataSource": {
     stepType: "queryDataSource",
-    description: "Search a vector data source (RAG) and return relevant document chunks.",
+    description: "Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Search a vector data source (RAG) and return relevant document chunks.",
     usageNotes: "- Queries a vectorized data source and returns the most relevant chunks.\n- Useful for retrieval-augmented generation (RAG) workflows.",
     inputSchema: { "type": "object", "properties": { "dataSourceId": { "type": "string", "description": "ID of the vector data source to query" }, "query": { "type": "string", "description": "The search query to run against the data source" }, "maxResults": { "type": "number", "description": "Maximum number of chunks to return (recommended 1-3)" } }, "required": ["dataSourceId", "query", "maxResults"] },
     outputSchema: { "type": "object", "properties": { "text": { "type": "string", "description": "All matching chunks joined with newlines" }, "chunks": { "type": "array", "items": { "type": "string" }, "description": "Individual matching text chunks from the data source" }, "query": { "type": "string", "description": "The resolved search query that was executed" }, "citations": { "type": "array", "items": {}, "description": "Source citations for the matched chunks" }, "latencyMs": { "type": "number", "description": "Query execution time in milliseconds" } }, "required": ["text", "chunks", "query", "citations", "latencyMs"] }
@@ -3380,7 +3380,7 @@ var stepMetadata = {
   },
   "uploadDataSourceDocument": {
     stepType: "uploadDataSourceDocument",
-    description: "Upload a file into an existing data source from a URL or raw text content.",
+    description: "Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from @mindstudio-ai/agent instead, which owns parsing, chunking, embedding and citations. Upload a file into an existing data source from a URL or raw text content.",
     usageNotes: '- If "file" is a single URL, the file is downloaded from that URL and uploaded.\n- If "file" is any other string, a .txt document is created from that content and uploaded.\n- The block waits (polls) for processing to complete before transitioning, up to 5 minutes.\n- Once processing finishes, vectors are loaded into Milvus so the data source is immediately queryable.\n- Supported file types (when using a URL) are the same as the data source upload UI (PDF, DOCX, TXT, etc.).',
     inputSchema: { "type": "object", "properties": { "dataSourceId": { "type": "string", "description": "ID of the target data source (supports variable interpolation)" }, "file": { "type": "string", "description": "A URL to download, or raw text content to create a .txt document from (supports variable interpolation)" }, "fileName": { "type": "string", "description": "Display name for the document (supports variable interpolation)" } }, "required": ["dataSourceId", "file", "fileName"] },
     outputSchema: { "description": "This step does not produce output data." }