feat(v1): add public actions api

apps/api/src/controllers/v1/types.ts

@@ -1,7 +1,7 @@
 import { Request, Response } from "express";
 import { z } from "zod";
 import { isUrlBlocked } from "../../scraper/WebScraper/utils/blocklist";
-import { ExtractorOptions, PageOptions } from "../../lib/entities";
+import { Action, ExtractorOptions, PageOptions } from "../../lib/entities";
 import { protocolIncluded, checkUrl } from "../../lib/validateUrl";
 import { PlanType } from "../../types";
 
@@ -57,6 +57,21 @@ export const extractOptions = z.object({
 
 export type ExtractOptions = z.infer<typeof extractOptions>;
 
+export const actionsSchema = z.array(z.union([
+  z.object({
+    type: z.literal("wait"),
+    milliseconds: z.number().int().positive().finite(),
+  }),
+  z.object({
+    type: z.literal("click"),
+    selector: z.string(),
+  }),
+  z.object({
+    type: z.literal("screenshot"),
+    fullPage: z.boolean().default(false),
+  }),
+]));
+
 export const scrapeOptions = z.object({
   formats: z
     .enum([
@@ -80,6 +95,7 @@ export const scrapeOptions = z.object({
   waitFor: z.number().int().nonnegative().finite().safe().default(0),
   extract: extractOptions.optional(),
   parsePDF: z.boolean().default(true),
+  actions: actionsSchema.optional(),
 }).strict(strictMessage)
 
 
@@ -185,6 +201,9 @@ export type Document = {
   rawHtml?: string;
   links?: string[];
   screenshot?: string;
+  actions?: {
+    screenshots: string[];
+  };
   metadata: {
     title?: string;
     description?: string;
@@ -336,6 +355,7 @@ export function legacyScrapeOptions(x: ScrapeOptions): PageOptions {
     screenshot: x.formats.includes("screenshot"),
     fullPageScreenshot: x.formats.includes("screenshot@fullPage"),
     parsePDF: x.parsePDF,
+    actions: x.actions as Action[], // no strict null checking grrrr - mogery
   };
 }
 
@@ -370,6 +390,7 @@ export function legacyDocumentConverter(doc: any): Document {
     html: doc.html,
     extract: doc.llm_extraction,
     screenshot: doc.screenshot ?? doc.fullPageScreenshot,
+    actions: doc.actions ?? undefined,
     metadata: {
       ...doc.metadata,
       pageError: undefined,

apps/api/src/lib/entities.ts

@@ -110,6 +110,9 @@ export class Document {
   childrenLinks?: string[];
   provider?: string;
   warning?: string;
+  actions?: {
+    screenshots: string[];
+  }
 
   index?: number;
   linksOnPage?: string[]; // Add this new field as a separate property
@@ -149,7 +152,7 @@ export class SearchResult {
 
 export interface FireEngineResponse {
   html: string;
-  screenshot: string;
+  screenshots?: string[];
   pageStatusCode?: number;
   pageError?: string;
 }

apps/api/src/scraper/WebScraper/scrapers/fireEngine.ts

@@ -136,7 +136,7 @@ export async function scrapWithFireEngine({
       
       Logger.debug(`⛏️ Fire-Engine (${engine}): Request timed out for ${url}`);
       logParams.error_message = "Request timed out";
-      return { html: "", screenshot: "", pageStatusCode: null, pageError: "" };
+      return { html: "", pageStatusCode: null, pageError: "" };
     }
 
     if (checkStatusResponse.status !== 200 || checkStatusResponse.data.error) {
@@ -155,7 +155,6 @@ export async function scrapWithFireEngine({
 
       return {
         html: "",
-        screenshot: "",
         pageStatusCode,
         pageError: checkStatusResponse.data?.pageError ?? checkStatusResponse.data?.error,
       };
@@ -171,7 +170,7 @@ export async function scrapWithFireEngine({
       logParams.success = true;
       logParams.response_code = pageStatusCode;
       logParams.error_message = pageError;
-      return { html: content, screenshot: "", pageStatusCode, pageError };
+      return { html: content, pageStatusCode, pageError };
     } else {
       const data = checkStatusResponse.data;
       
@@ -183,7 +182,7 @@ export async function scrapWithFireEngine({
       logParams.error_message = data.pageError ?? data.error;
       return {
         html: data.content ?? "",
-        screenshot: data.screenshot ?? "",
+        screenshots: data.screenshots,
         pageStatusCode: data.pageStatusCode,
         pageError: data.pageError ?? data.error,
       };
@@ -196,7 +195,7 @@ export async function scrapWithFireEngine({
       Logger.debug(`⛏️ Fire-Engine: Failed to fetch url: ${url} | Error: ${error}`);
       logParams.error_message = error.message || error;
     }
-    return { html: "", screenshot: "", pageStatusCode: null, pageError: logParams.error_message };
+    return { html: "", pageStatusCode: null, pageError: logParams.error_message };
   } finally {
     const endTime = Date.now();
     logParams.time_taken_seconds = (endTime - logParams.startTime) / 1000;

apps/api/src/scraper/WebScraper/single_url.ts

@@ -69,8 +69,13 @@ function getScrapingFallbackOrder(
   defaultScraper?: string,
   isWaitPresent: boolean = false,
   isScreenshotPresent: boolean = false,
-  isHeadersPresent: boolean = false
+  isHeadersPresent: boolean = false,
+  isActionsPresent: boolean = false,
 ) {
+  if (isActionsPresent) {
+    return useFireEngine ? ["fire-engine;chrome-cdp"] : [];
+  }
+
   const availableScrapers = baseScrapers.filter((scraper) => {
     switch (scraper) {
       case "scrapingBee":
@@ -170,6 +175,9 @@ export async function scrapSingleUrl(
     let scraperResponse: {
       text: string;
       screenshot: string;
+      actions?: {
+        screenshots: string[];
+      };
       metadata: { pageStatusCode?: number; pageError?: string | null };
     } = { text: "", screenshot: "", metadata: {} };
     let screenshot = "";
@@ -217,7 +225,14 @@ export async function scrapSingleUrl(
             teamId,
           });
           scraperResponse.text = response.html;
-          scraperResponse.screenshot = response.screenshot;
+          if (pageOptions.screenshot || pageOptions.fullPageScreenshot) {
+            scraperResponse.screenshot = (response.screenshots ?? []).splice(0, 1)[0] ?? "";
+          }
+          if (pageOptions.actions) {
+            scraperResponse.actions = {
+              screenshots: response.screenshots ?? [],
+            };
+          }
           scraperResponse.metadata.pageStatusCode = response.pageStatusCode;
           scraperResponse.metadata.pageError = response.pageError;
         }
@@ -283,9 +298,6 @@ export async function scrapSingleUrl(
             ]) : ([]),
             pageOptions: customScraperResult.pageOptions,
           });
-          if (screenshot) {
-            customScrapedContent.screenshot = screenshot;
-          }
           break;
         case "pdf":
           const { content, pageStatusCode, pageError } =
@@ -295,7 +307,6 @@ export async function scrapSingleUrl(
             );
           customScrapedContent = {
             html: content,
-            screenshot,
             pageStatusCode,
             pageError,
           };
@@ -305,7 +316,6 @@ export async function scrapSingleUrl(
 
     if (customScrapedContent) {
       scraperResponse.text = customScrapedContent.html;
-      screenshot = customScrapedContent.screenshot;
     }
     //* TODO: add an optional to return markdown or structured/extracted content
     let cleanedHtml = removeUnwantedElements(scraperResponse.text, pageOptions);
@@ -325,16 +335,18 @@ export async function scrapSingleUrl(
       html: cleanedHtml,
       rawHtml: scraperResponse.text,
       screenshot: scraperResponse.screenshot,
+      actions: scraperResponse.actions,
       pageStatusCode: scraperResponse.metadata.pageStatusCode,
       pageError: scraperResponse.metadata.pageError || undefined,
     };
   };
 
-  let { text, html, rawHtml, screenshot, pageStatusCode, pageError } = {
+  let { text, html, rawHtml, screenshot, actions, pageStatusCode, pageError } = {
     text: "",
     html: "",
     rawHtml: "",
     screenshot: "",
+    actions: undefined,
     pageStatusCode: 200,
     pageError: undefined,
   };
@@ -350,7 +362,8 @@ export async function scrapSingleUrl(
       defaultScraper,
       pageOptions && pageOptions.waitFor && pageOptions.waitFor > 0,
       pageOptions && (pageOptions.screenshot || pageOptions.fullPageScreenshot) && (pageOptions.screenshot === true || pageOptions.fullPageScreenshot === true),
-      pageOptions && pageOptions.headers && pageOptions.headers !== undefined
+      pageOptions && pageOptions.headers && pageOptions.headers !== undefined,
+      pageOptions && Array.isArray(pageOptions.actions) && pageOptions.actions.length > 0,
     );
 
     for (const scraper of scrapersInOrder) {
@@ -367,6 +380,7 @@ export async function scrapSingleUrl(
       html = attempt.html ?? "";
       rawHtml = attempt.rawHtml ?? "";
       screenshot = attempt.screenshot ?? "";
+      actions = attempt.actions ?? undefined;
 
       if (attempt.pageStatusCode) {
         pageStatusCode = attempt.pageStatusCode;
@@ -404,45 +418,27 @@ export async function scrapSingleUrl(
       linksOnPage = extractLinks(rawHtml, urlToScrap);
     }
 
-    let document: Document;
+    let document: Document = {
-    if (screenshot && screenshot.length > 0) {
+      content: text,
-      document = {
+      markdown: pageOptions.includeMarkdown || pageOptions.includeExtract ? text : undefined,
-        content: text,
+      html: pageOptions.includeHtml ? html : undefined,
-        markdown: pageOptions.includeMarkdown || pageOptions.includeExtract ? text : undefined,
+      rawHtml:
-        html: pageOptions.includeHtml ? html : undefined,
+        pageOptions.includeRawHtml ||
-        rawHtml:
+          (extractorOptions?.mode === "llm-extraction-from-raw-html" && pageOptions.includeExtract)
-          pageOptions.includeRawHtml ||
+          ? rawHtml
-            (extractorOptions?.mode === "llm-extraction-from-raw-html" && pageOptions.includeExtract)
+          : undefined,
-            ? rawHtml
+      linksOnPage: pageOptions.includeLinks ? linksOnPage : undefined,
-            : undefined,
+      actions,
-        linksOnPage: pageOptions.includeLinks ? linksOnPage : undefined,
+      metadata: {
-        metadata: {
+        ...metadata,
-          ...metadata,
+        ...(screenshot && screenshot.length > 0 ? ({
-          screenshot: screenshot,
+          screenshot,
-          sourceURL: urlToScrap,
+        }) : {}),
-          pageStatusCode: pageStatusCode,
+        sourceURL: urlToScrap,
-          pageError: pageError,
+        pageStatusCode: pageStatusCode,
-        },
+        pageError: pageError,
-      };
+      },
-    } else {
+    };
-      document = {
-        content: text,
-        markdown: pageOptions.includeMarkdown || pageOptions.includeExtract ? text : undefined,
-        html: pageOptions.includeHtml ? html : undefined,
-        rawHtml:
-          pageOptions.includeRawHtml ||
-            (extractorOptions?.mode === "llm-extraction-from-raw-html" && pageOptions.includeExtract)
-            ? rawHtml
-            : undefined,
-        metadata: {
-          ...metadata,
-          sourceURL: urlToScrap,
-          pageStatusCode: pageStatusCode,
-          pageError: pageError,
-        },
-        linksOnPage: pageOptions.includeLinks ? linksOnPage : undefined,
-      };
-    }
 
     return document;
   } catch (error) {