← All use cases

中文

Sales & marketing · Claude Cowork

Score people who engage with a LinkedIn post and draft messages to the strongest fits

Idea + starter prompt, not run by us yet

For: Founders and marketers who post on LinkedIn and want to follow up with the right people

Pietro hooks a scraping tool and a contact data tool up to Claude, and gives Cowork a folder describing his company, ideal customer and offer. Given a post link, it pulls the commenters, fills in their details, scores each from 0 to 5 for fit and drafts messages for the 4s and 5s. He saves it as a skill.

Steps

  1. Make a folder with notes on your company, ideal customer and offer.
  2. Connect the data tools and check their credit costs first.
  3. Paste the starter prompt with one post link.
  4. Check the scores on a few people yourself and send messages by hand.
  5. Save the whole skill folder, including the scoring rules.

Starter prompt

Starter prompt. Change the [brackets] first.

Read the notes in [folder] about my company, ideal customer and offer.

For the LinkedIn post at [post link], get the list of people who commented. For each person, add their role and company, then score fit from 0 to 5 with one line of reason.

Before spending any data credits, tell me how many lookups you plan. Show me the scored list. Then draft a short message only for people who scored 4 or 5. Do not send any message.

Original post: Substack · found via agentusecases.ai. Written up in our own words. We haven't run it, so test it on made-up data first.