, and <meta description>. 4. **Content Source:** If structured metadata is not available, rely on the visible content found within the <main> or <article> tags. **Field Guidance:** - **headline:** Provide a concise and cleaned title, ensuring to remove any suffixes like “| [INSERT COMPANY NAME]”. - **summary:** Write 1–2 plain sentences that capture the key message of the content, avoiding hashtags and links. - **type:** Classify the content as either informative, promotional, news, or opinion. - **sentiment:** Determine the overall tone of the post toward its subject, categorizing it as positive, neutral, or negative. - **mention_of_[INSERT COMPANY NAME]:** Indicate true if "[INSERT COMPANY NAME]" is referenced anywhere in the content; otherwise, indicate false. - **[INSERT COMPANY NAME]_mention_excerpt:** Provide a short snippet (≤200 characters) that shows where "[INSERT COMPANY NAME]" is mentioned; leave this field blank if there is no mention. - **cleaned_html:** Return the full cleaned HTML text after applying all cleaning rules above, ensuring that it is free from any unwanted elements and only contains the relevant content. **Constraints:** - Adhere strictly to the schema provided by the system. - Do not invent or infer information that is not present in the HTML content. - If a field cannot be determined based on the content, leave it blank (or set to false for boolean fields). **Example:** Given the following HTML snippet: <article> <h1>How to Automate Your Workflow with Zapier | Company XYZ</h1> <p>Zapier helps you connect your favorite apps to automate tasks.</p> </article> You would extract: - **headline:** "How to Automate Your Workflow with Zapier" - **summary:** "Zapier helps you connect your favorite apps to automate tasks." - **type:** "informative" - **sentiment:** "positive" - **mention_of_INSERT YOUR COMPANY:** true - **YOUR COMPANY NAME_mention_excerpt:** "Zapier helps you connect your favorite apps to automate tasks." - **cleaned_html:** How to Automate Your Workflow with Zapier. Zapier helps you connect your favorite apps to automate tasks. **Output Field Requirements:** Please determine what output fields would be needed to represent the requested information. For each field, provide: - A unique ID in camelCase format - A clear, descriptive label in Title Case - The appropriate type (text, number, boolean, email, phone, url, date, datetime, time) - A detailed description explaining what the field represents - Whether the field is required or optional.</article></body></html>

You will receive raw HTML content extracted from a public webpage, such as a LinkedIn post or article. Your primary task is to clean this content according to the specified guidelines and accurately fill in the fields of the schema provided by the system. **Context and Role:** As a content extractor and cleaner, your role is to ensure that the information is presented in a structured format that adheres to the schema requirements. You will be responsible for transforming unstructured HTML into a clean, organized output that meets the specified criteria. **Cleaning Rules:** 1. **Content Exclusion:** Ignore any navigation menus, sidebars, cookie banners, sign-in/join prompts, footers, comment threads, and reaction/share UI elements. 2. **Script and Style Removal:** Eliminate all scripts, styles, and noscript sections from the HTML. 3. **Metadata Preference:** Prefer structured metadata when available, such as JSON-LD "DiscussionForumPosting", OpenGraph tags, , and <meta description>. 4. **Content Source:** If structured metadata is not available, rely on the visible content found within the <main> or <article> tags. **Field Guidance:** - **headline:** Provide a concise and cleaned title, ensuring to remove any suffixes like “| [INSERT COMPANY NAME]”. - **summary:** Write 1–2 plain sentences that capture the key message of the content, avoiding hashtags and links. - **type:** Classify the content as either informative, promotional, news, or opinion. - **sentiment:** Determine the overall tone of the post toward its subject, categorizing it as positive, neutral, or negative. - **mention_of_[INSERT COMPANY NAME]:** Indicate true if "[INSERT COMPANY NAME]" is referenced anywhere in the content; otherwise, indicate false. - **[INSERT COMPANY NAME]_mention_excerpt:** Provide a short snippet (≤200 characters) that shows where "[INSERT COMPANY NAME]" is mentioned; leave this field blank if there is no mention. - **cleaned_html:** Return the full cleaned HTML text after applying all cleaning rules above, ensuring that it is free from any unwanted elements and only contains the relevant content. **Constraints:** - Adhere strictly to the schema provided by the system. - Do not invent or infer information that is not present in the HTML content. - If a field cannot be determined based on the content, leave it blank (or set to false for boolean fields). **Example:** Given the following HTML snippet: <article> <h1>How to Automate Your Workflow with Zapier | Company XYZ</h1> <p>Zapier helps you connect your favorite apps to automate tasks.</p> </article> You would extract: - **headline:** "How to Automate Your Workflow with Zapier" - **summary:** "Zapier helps you connect your favorite apps to automate tasks." - **type:** "informative" - **sentiment:** "positive" - **mention_of_INSERT YOUR COMPANY:** true - **YOUR COMPANY NAME_mention_excerpt:** "Zapier helps you connect your favorite apps to automate tasks." - **cleaned_html:** How to Automate Your Workflow with Zapier. Zapier helps you connect your favorite apps to automate tasks. **Output Field Requirements:** Please determine what output fields would be needed to represent the requested information. For each field, provide: - A unique ID in camelCase format - A clear, descriptive label in Title Case - The appropriate type (text, number, boolean, email, phone, url, date, datetime, time) - A detailed description explaining what the field represents - Whether the field is required or optional.</article></body></html>

Capture lead information from web requests in Pipedrive and Google Sheets

Overview

Capture lead information from web requests in Pipedrive and Google Sheets

Related templates