@mindstudio-ai/agent 0.1.78 → 0.1.79

package/dist/index.d.ts

@@ -1979,8 +1979,10 @@ interface AirtableGetTableRecordsStepOutput {
 interface AnalyzeImageStepInput {
     /** Instructions describing what to look for or extract from the image */
     prompt: string;
-    /** URL of the image to analyze */
-    imageUrl: string;
+    /** URL of a single image to analyze. Kept for backward compatibility; prefer imageUrls. */
+    imageUrl?: string;
+    /** One or more image URLs to analyze together in a single model request */
+    imageUrls?: string[];
     /** Optional model configuration override. Uses the workflow's default vision model if not specified */
     visionModelOverride?: {
         model: string;
@@ -6043,12 +6045,13 @@ interface StepMethods {
      * @remarks
      * - Uses the configured vision model to generate a text analysis of the image.
      * - The prompt should describe what to look for or extract from the image.
+     * - Pass imageUrl for a single image, or imageUrls for multiple images analyzed together in one request.
+     * - Most vision models (OpenAI, Grok, Gemini) accept multiple images in one request. Ideogram describe is single-image only.
      *
      * @example
      * ```typescript
      * const result = await agent.analyzeImage({
      *   prompt: ``,
-     *   imageUrl: ``,
      * });
      * ```
      */

package/dist/index.js

@@ -1978,8 +1978,8 @@ var stepMetadata = {
   "analyzeImage": {
     stepType: "analyzeImage",
     description: "Analyze an image using a vision model based on a text prompt.",
-    usageNotes: "- Uses the configured vision model to generate a text analysis of the image.\n- The prompt should describe what to look for or extract from the image.",
-    inputSchema: { "type": "object", "properties": { "prompt": { "type": "string", "description": "Instructions describing what to look for or extract from the image" }, "imageUrl": { "type": "string", "description": "URL of the image to analyze" }, "visionModelOverride": { "anyOf": [{ "type": "object", "properties": { "model": { "type": "string" }, "config": { "type": "object", "properties": {}, "required": [] } }, "required": ["model"] }, { "type": "object", "properties": { "model": { "type": "string", "description": 'Model identifier (e.g. "gpt-4", "claude-3-opus")' }, "temperature": { "type": "number", "description": "Sampling temperature for the model (0-2)" }, "maxResponseTokens": { "type": "number", "description": "Maximum number of tokens in the model's response" }, "ignorePreamble": { "type": "boolean", "description": "Whether to skip the system preamble/instructions" }, "userMessagePreprocessor": { "type": "object", "properties": { "dataSource": { "type": "string", "description": "Data source identifier for the preprocessor" }, "messageTemplate": { "type": "string", "description": "Template string applied to user messages before sending to the model" }, "maxResults": { "type": "number", "description": "Maximum number of results to include from the data source" }, "enabled": { "type": "boolean", "description": "Whether the preprocessor is active" }, "shouldInherit": { "type": "boolean", "description": "Whether child steps should inherit this preprocessor configuration" } }, "description": "Preprocessor applied to user messages before sending to the model" }, "preamble": { "type": "string", "description": "System preamble/instructions for the model" }, "multiModelEnabled": { "type": "boolean", "description": "Whether multi-model candidate generation is enabled" }, "editResponseEnabled": { "type": "boolean", "description": "Whether the user can edit the model's response" }, "config": { "type": "object", "properties": {}, "required": [], "description": "Additional model-specific configuration" } }, "required": ["model", "temperature", "maxResponseTokens"] }] } }, "required": ["prompt", "imageUrl"] },
+    usageNotes: "- Uses the configured vision model to generate a text analysis of the image.\n- The prompt should describe what to look for or extract from the image.\n- Pass imageUrl for a single image, or imageUrls for multiple images analyzed together in one request.\n- Most vision models (OpenAI, Grok, Gemini) accept multiple images in one request. Ideogram describe is single-image only.",
+    inputSchema: { "type": "object", "properties": { "prompt": { "type": "string", "description": "Instructions describing what to look for or extract from the image" }, "imageUrl": { "type": "string", "description": "URL of a single image to analyze. Kept for backward compatibility; prefer imageUrls." }, "imageUrls": { "type": "array", "items": { "type": "string" }, "description": "One or more image URLs to analyze together in a single model request" }, "visionModelOverride": { "anyOf": [{ "type": "object", "properties": { "model": { "type": "string" }, "config": { "type": "object", "properties": {}, "required": [] } }, "required": ["model"] }, { "type": "object", "properties": { "model": { "type": "string", "description": 'Model identifier (e.g. "gpt-4", "claude-3-opus")' }, "temperature": { "type": "number", "description": "Sampling temperature for the model (0-2)" }, "maxResponseTokens": { "type": "number", "description": "Maximum number of tokens in the model's response" }, "ignorePreamble": { "type": "boolean", "description": "Whether to skip the system preamble/instructions" }, "userMessagePreprocessor": { "type": "object", "properties": { "dataSource": { "type": "string", "description": "Data source identifier for the preprocessor" }, "messageTemplate": { "type": "string", "description": "Template string applied to user messages before sending to the model" }, "maxResults": { "type": "number", "description": "Maximum number of results to include from the data source" }, "enabled": { "type": "boolean", "description": "Whether the preprocessor is active" }, "shouldInherit": { "type": "boolean", "description": "Whether child steps should inherit this preprocessor configuration" } }, "description": "Preprocessor applied to user messages before sending to the model" }, "preamble": { "type": "string", "description": "System preamble/instructions for the model" }, "multiModelEnabled": { "type": "boolean", "description": "Whether multi-model candidate generation is enabled" }, "editResponseEnabled": { "type": "boolean", "description": "Whether the user can edit the model's response" }, "config": { "type": "object", "properties": {}, "required": [], "description": "Additional model-specific configuration" } }, "required": ["model", "temperature", "maxResponseTokens"] }] } }, "required": ["prompt"] },
     outputSchema: { "type": "object", "properties": { "analysis": { "type": "string", "description": "Text analysis of the image generated by the vision model" } }, "required": ["analysis"] }
   },
   "analyzeVideo": {
@@ -5212,7 +5212,7 @@ var monacoSnippets = {
   "airtableDeleteRecord": { fields: [["baseId", "string"], ["tableId", "string"], ["recordId", "string"]], outputKeys: ["deleted"] },
   "airtableGetRecord": { fields: [["baseId", "string"], ["tableId", "string"], ["recordId", "string"]], outputKeys: ["record"] },
   "airtableGetTableRecords": { fields: [["baseId", "string"], ["tableId", "string"]], outputKeys: ["records"] },
-  "analyzeImage": { fields: [["prompt", "string"], ["imageUrl", "string"]], outputKeys: ["analysis"] },
+  "analyzeImage": { fields: [["prompt", "string"]], outputKeys: ["analysis"] },
   "analyzeVideo": { fields: [["prompt", "string"], ["videoUrl", "string"]], outputKeys: ["analysis"] },
   "captureThumbnail": { fields: [["videoUrl", "string"], ["at", "string"]], outputKeys: ["thumbnailUrl"] },
   "checkAppRole": { fields: [["roleName", "string"]], outputKeys: ["hasRole", "userRoles"] },