From fabbd0aa099df82b492c1a02045a48d3ceb29661 Mon Sep 17 00:00:00 2001
From: Sweets Sweetman <sweetmantech@gmail.com>
Date: Thu, 2 Jul 2026 13:49:09 -0500
Subject: [PATCH] feat(apify): LinkedIn scrape returns posts when a posts depth
 is requested
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A requested posts depth switches the LinkedIn scrape from
harvestapi/linkedin-profile-scraper (whose dataset has zero post fields)
to harvestapi/linkedin-profile-posts with {targetUrls, maxPosts} — post
items with engagement counts (likes, comments, shares, reaction
breakdown). posts omitted keeps the profile actor with byte-identical
input. Same handle guards, URL normalization, webhooks, and error paths.

Matches X/YouTube semantics (depth → post items; omitted → snapshot).
Actor verified live 2026-07-02: run aKsHPucJ3thxuosSd, 5 posts @ $0.01005
($0.002/post) — the shipped 5 + posts credit rule covers it at ~5x margin,
no billing change.

Implements recoupable/chat#1838; contract in recoupable/docs#260.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---
 .../startLinkedinProfileScraping.test.ts      | 28 +++++++++++++++++++
 .../linkedin/startLinkedinProfileScraping.ts  | 25 +++++++++++++----
 2 files changed, 47 insertions(+), 6 deletions(-)

diff --git a/lib/apify/__tests__/startLinkedinProfileScraping.test.ts b/lib/apify/__tests__/startLinkedinProfileScraping.test.ts
index dcd48700..07c05dc5 100644
--- a/lib/apify/__tests__/startLinkedinProfileScraping.test.ts
+++ b/lib/apify/__tests__/startLinkedinProfileScraping.test.ts
@@ -35,4 +35,32 @@ describe("startLinkedinProfileScraping", () => {
     );
     expect(start).not.toHaveBeenCalled();
   });
+
+  it("runs the posts actor with maxPosts when a posts depth is requested", async () => {
+    const r = await startLinkedinProfileScraping("sweetmaneth", 20);
+    expect(apifyClient.actor).toHaveBeenCalledWith("harvestapi/linkedin-profile-posts");
+    expect(start).toHaveBeenCalledWith(
+      { targetUrls: ["https://www.linkedin.com/in/sweetmaneth"], maxPosts: 20 },
+      { webhooks: expect.any(Array) },
+    );
+    expect(r).toEqual({ runId: "run-1", datasetId: "ds-1" });
+  });
+
+  it("posts depth keeps the full-URL passthrough and the handle guards", async () => {
+    await startLinkedinProfileScraping("https://www.linkedin.com/in/drew-thurlow", 5);
+    expect(start).toHaveBeenCalledWith(
+      { targetUrls: ["https://www.linkedin.com/in/drew-thurlow"], maxPosts: 5 },
+      { webhooks: expect.any(Array) },
+    );
+    await expect(startLinkedinProfileScraping("in", 5)).rejects.toThrow(/Invalid LinkedIn handle/);
+  });
+
+  it("posts omitted keeps the legacy profile actor and input byte-identical", async () => {
+    await startLinkedinProfileScraping("sweetmaneth", undefined);
+    expect(apifyClient.actor).toHaveBeenCalledWith("harvestapi/linkedin-profile-scraper");
+    expect(start).toHaveBeenCalledWith(
+      { urls: ["https://www.linkedin.com/in/sweetmaneth"] },
+      { webhooks: expect.any(Array) },
+    );
+  });
 });
diff --git a/lib/apify/linkedin/startLinkedinProfileScraping.ts b/lib/apify/linkedin/startLinkedinProfileScraping.ts
index ff323194..d9752d0d 100644
--- a/lib/apify/linkedin/startLinkedinProfileScraping.ts
+++ b/lib/apify/linkedin/startLinkedinProfileScraping.ts
@@ -4,14 +4,24 @@ import { OUTSTANDING_ERROR } from "@/lib/apify/errors";
 import { ApifyRunInfo } from "@/lib/apify/types";
 
 /**
- * Starts a LinkedIn profile scrape via the harvestapi actor.
+ * Starts a LinkedIn scrape via the harvestapi actors.
  * Mirrors the other `start<Platform>ProfileScraping` modules.
  *
+ * The profile actor's dataset has no post fields, so a requested `posts`
+ * depth switches to the posts actor (post items with engagement counts) —
+ * the same semantics as X/YouTube: depth requested → post items in the
+ * dataset; omitted → the legacy profile snapshot.
+ *
  * @param handle - A LinkedIn vanity slug (e.g. `drew-thurlow`) or full profile URL.
+ * @param posts - Max recent posts to scrape; omitted → profile snapshot.
  * @returns Apify run info, or null if the run failed to start.
  * @see https://apify.com/harvestapi/linkedin-profile-scraper
+ * @see https://apify.com/harvestapi/linkedin-profile-posts
  */
-const startLinkedinProfileScraping = async (handle: string): Promise<ApifyRunInfo | null> => {
+const startLinkedinProfileScraping = async (
+  handle: string,
+  posts?: number,
+): Promise<ApifyRunInfo | null> => {
   const cleanHandle = handle.trim().replace(/^@/, "");
   // Legacy rows stored the LinkedIn path prefix ("in") as the username — never
   // scrape those (linkedin.com/in/in is a real, wrong profile URL); fail loudly.
@@ -23,12 +33,15 @@ const startLinkedinProfileScraping = async (handle: string): Promise<ApifyRunInf
     ? cleanHandle
     : `https://www.linkedin.com/in/${cleanHandle}`;
 
-  const input = { urls: [targetUrl] };
+  const actorId =
+    posts === undefined
+      ? "harvestapi/linkedin-profile-scraper"
+      : "harvestapi/linkedin-profile-posts";
+  const input =
+    posts === undefined ? { urls: [targetUrl] } : { targetUrls: [targetUrl], maxPosts: posts };
 
   try {
-    const run = await apifyClient
-      .actor("harvestapi/linkedin-profile-scraper")
-      .start(input, { webhooks: getApifyWebhooks() });
+    const run = await apifyClient.actor(actorId).start(input, { webhooks: getApifyWebhooks() });
 
     if (!run?.id || !run?.defaultDatasetId) {
       console.error("Failed to start LinkedIn profile scraping for handle:", handle);