@credal/actions 0.2.117 → 0.2.118

package/dist/actions/autogen/templates.js

@@ -3208,6 +3208,29 @@ export const firecrawlScrapeUrlDefinition = {
                 description: "Optional wait time in milliseconds before scraping the page",
                 minimum: 0,
             },
+            onlyMainContent: {
+                type: "boolean",
+                description: "Extract only the main content of the page, excluding headers, footers, and navigation",
+            },
+            formats: {
+                type: "array",
+                description: "Array of formats to return",
+                items: {
+                    type: "string",
+                    enum: [
+                        "content",
+                        "json",
+                        "html",
+                        "screenshot",
+                        "markdown",
+                        "rawHtml",
+                        "links",
+                        "screenshot@fullPage",
+                        "extract",
+                        "changeTracking",
+                    ],
+                },
+            },
         },
     },
     output: {

package/dist/actions/autogen/types.d.ts

@@ -3142,12 +3142,18 @@ export type firecrawlDeepResearchFunction = ActionFunction<firecrawlDeepResearch
 export declare const firecrawlScrapeUrlParamsSchema: z.ZodObject<{
     url: z.ZodString;
     waitMs: z.ZodOptional<z.ZodNumber>;
+    onlyMainContent: z.ZodOptional<z.ZodBoolean>;
+    formats: z.ZodOptional<z.ZodArray<z.ZodEnum<["content", "json", "html", "screenshot", "markdown", "rawHtml", "links", "screenshot@fullPage", "extract", "changeTracking"]>, "many">>;
 }, "strip", z.ZodTypeAny, {
     url: string;
     waitMs?: number | undefined;
+    onlyMainContent?: boolean | undefined;
+    formats?: ("content" | "json" | "html" | "screenshot" | "markdown" | "rawHtml" | "links" | "screenshot@fullPage" | "extract" | "changeTracking")[] | undefined;
 }, {
     url: string;
     waitMs?: number | undefined;
+    onlyMainContent?: boolean | undefined;
+    formats?: ("content" | "json" | "html" | "screenshot" | "markdown" | "rawHtml" | "links" | "screenshot@fullPage" | "extract" | "changeTracking")[] | undefined;
 }>;
 export type firecrawlScrapeUrlParamsType = z.infer<typeof firecrawlScrapeUrlParamsSchema>;
 export declare const firecrawlScrapeUrlOutputSchema: z.ZodObject<{

package/dist/actions/autogen/types.js

@@ -979,6 +979,25 @@ export const firecrawlDeepResearchOutputSchema = z.object({
 export const firecrawlScrapeUrlParamsSchema = z.object({
     url: z.string().describe("The URL to scrape"),
     waitMs: z.number().gte(0).describe("Optional wait time in milliseconds before scraping the page").optional(),
+    onlyMainContent: z
+        .boolean()
+        .describe("Extract only the main content of the page, excluding headers, footers, and navigation")
+        .optional(),
+    formats: z
+        .array(z.enum([
+        "content",
+        "json",
+        "html",
+        "screenshot",
+        "markdown",
+        "rawHtml",
+        "links",
+        "screenshot@fullPage",
+        "extract",
+        "changeTracking",
+    ]))
+        .describe("Array of formats to return")
+        .optional(),
 });
 export const firecrawlScrapeUrlOutputSchema = z.object({ content: z.string().describe("The content of the URL") });
 export const firecrawlSearchAndScrapeParamsSchema = z.object({

package/dist/actions/providers/firecrawl/scrapeUrl.js

@@ -13,11 +13,70 @@ const scrapeUrl = (_a) => __awaiter(void 0, [_a], void 0, function* ({ params, a
     const firecrawl = new FirecrawlApp({
         apiKey: authParams.apiKey,
     });
-    const result = yield firecrawl.scrapeUrl(params.url, Object.assign({}, (params.waitMs !== undefined && {
+    const result = yield firecrawl.scrapeUrl(params.url, Object.assign(Object.assign(Object.assign({}, (params.waitMs !== undefined && {
         actions: [{ type: "wait", milliseconds: params.waitMs }],
+    })), (params.onlyMainContent !== undefined && {
+        onlyMainContent: params.onlyMainContent,
+    })), (params.formats !== undefined &&
+        params.formats.length > 0 && {
+        formats: params.formats,
     })));
+    console.log("Result is: ", result);
+    if (!result.success) {
+        return firecrawlScrapeUrlOutputSchema.parse({
+            content: "",
+        });
+    }
+    // Extract content based on requested formats
+    let content = "";
+    if (params.formats && params.formats.length > 0) {
+        const contentParts = [];
+        for (const format of params.formats) {
+            let formatContent = undefined;
+            // Handle different format mappings
+            switch (format) {
+                case "rawHtml":
+                    formatContent = result.rawHtml;
+                    break;
+                case "markdown":
+                    formatContent = result.markdown;
+                    break;
+                case "html":
+                    formatContent = result.html;
+                    break;
+                case "links":
+                    formatContent = Array.isArray(result.links)
+                        ? result.links.map(link => (typeof link === "string" ? link : JSON.stringify(link))).join("\n")
+                        : JSON.stringify(result.links);
+                    break;
+                case "json":
+                    formatContent = result.json ? JSON.stringify(result.json, null, 2) : undefined;
+                    break;
+                case "extract":
+                    formatContent = result.extract ? JSON.stringify(result.extract, null, 2) : undefined;
+                    break;
+                case "screenshot":
+                    formatContent = result.screenshot;
+                    break;
+                case "changeTracking":
+                    formatContent = result.changeTracking ? JSON.stringify(result.changeTracking, null, 2) : undefined;
+                    break;
+                default:
+                    formatContent = result[format];
+            }
+            if (formatContent) {
+                const formatHeader = `=== ${format.toUpperCase()} ===`;
+                contentParts.push(`${formatHeader}\n${formatContent}`);
+            }
+        }
+        content = contentParts.join("\n\n");
+    }
+    else {
+        // Default to markdown if no formats specified
+        content = result.markdown || "";
+    }
     return firecrawlScrapeUrlOutputSchema.parse({
-        content: result.success ? result.markdown : "",
+        content,
     });
 });
 export default scrapeUrl;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@credal/actions",
-  "version": "0.2.117",
+  "version": "0.2.118",
   "type": "module",
   "description": "AI Actions by Credal AI",
   "sideEffects": false,