post-engagers（岗位参与者）

LinkedIn Post Engagers

Turn LinkedIn post engagement into a prospecting list. Extract commenters, reactors, and reposters from any LinkedIn post — then enrich and upload to Extruct for outreach.

Related Skills

post-engagers → email-search → email-generation → campaign-sending

This skill produces a people table. The next step (email-search) gets verified emails, then email-generation drafts personalized outreach.

Extruct API Operations

This skill delegates all Extruct API calls to the extruct-api skill.

For all Extruct API operations, read and follow the instructions in skills/extruct-api/SKILL.md.

Table creation, row uploads, and data fetching are handled by the extruct-api skill. This skill focuses on scraping LinkedIn engagers and preparing the data — the extruct-api skill handles the API execution.

Inputs

LinkedIn Scraping Providers

This skill does not mandate a specific provider. Ask the user which LinkedIn scraping tool they want to use. Below are known options — the user may have others.

If the user doesn't know where to start:

• Anysite MCP is the simplest if they have it connected — no extra credentials needed
• Apify is a good general choice with pay-per-use pricing
• RapidAPI has multiple scrapers with free tiers

Workflow

Step 1: Collect post URLs and choose provider

1. Get the LinkedIn post URL(s) from the user. Accept one or multiple.
2. Ask which engagement types to scrape: comments, reactions, reposts, or all three.
3. Ask which LinkedIn scraping provider they want to use (see table above).
4. If the provider requires credentials, confirm they're available.

Extract the activity URN from each post URL. The numeric ID is typically after activity- or ugcPost- in the URL (e.g. activity:7433261939285385217).

Step 2: Scrape engagers

Use the chosen provider to fetch engagement data. The approach varies by provider:

If using Anysite MCP:

• Comments: mcp__claude_ai_Anysite__get_linkedin_post_comments with urn: "activity:{id}", count: 1500
• Reactions: mcp__claude_ai_Anysite__get_linkedin_post_reactions with urn: "activity:{id}", count: 1500
• Reposts: mcp__claude_ai_Anysite__get_linkedin_post_reposts with urn: "activity:{id}", count: 1500

If using another provider:

• Read or fetch the provider's API documentation
• Identify the endpoint, input format, and response structure
• Implement the scraping calls accordingly

For each engager, extract (field names vary by provider):

{
    "full_name": "...",
    "linkedin_url": "...",        # profile URL
    "headline": "...",            # job title / headline
    "engagement_type": "...",     # comment / reaction / repost
    "post_url": "...",            # which post they engaged with
}

If scraping multiple posts, tag each engager with the post_url they engaged with.

Step 3: Deduplicate and classify

1. Deduplicate by linkedin_url across all posts and engagement types. If someone both commented and reacted, keep both engagement types as a comma-separated value.
2. Apply the segment classifier to job titles (first match wins):

1. Present a segment breakdown to the user before proceeding:

Engager Summary:
- Total unique engagers: N
- Comments: N | Reactions: N | Reposts: N

Segment Breakdown:
  Founders / CEOs:         N (X%)
  Sales Leadership:        N (X%)
  Marketing Leadership:    N (X%)
  ...
  Other:                   N (X%)

1. Ask the user: "Want to filter to specific segments before uploading? (e.g. only Founders + Leadership)"

Step 4: Upload to Extruct people table

Create a new Extruct generic table or append to an existing one. Delegate to the extruct-api skill.

If creating a new table:

{
  "name": "{user-provided name or 'Post Engagers - {date}'}",
  "kind": "generic",
  "column_configs": [
    {"kind": "input", "name": "Full Name", "key": "full_name"},
    {"kind": "input", "name": "LinkedIn URL", "key": "linkedin_url"},
    {"kind": "input", "name": "Job Title", "key": "job_title"},
    {"kind": "input", "name": "Segment", "key": "segment"},
    {"kind": "input", "name": "Engagement Type", "key": "engagement_type"},
    {"kind": "input", "name": "Source Post", "key": "source_post"},
    {"kind": "input", "name": "Company", "key": "company"},
    {"kind": "input", "name": "Domain", "key": "domain"}
  ]
}

Upload rows in batches of 50 via the extruct-api skill.

If appending to an existing table:

• Fetch existing rows to deduplicate against current linkedin_url values
• Upload only new engagers

Step 5: Review and next steps

Present upload summary:

Upload Complete:
- Engagers uploaded: N
- Table: {table_name}
- URL: https://app.extruct.ai/tables/{table_id}

Segment Breakdown (uploaded):
  Founders / CEOs:     N
  Sales Leadership:    N
  ...

Suggest next steps:

• "Get emails" → run email-search on the people table to enrich with verified emails
• "Enrich companies" → run list-enrichment to add company data (industry, size, funding)
• "Draft outreach" → run email-generation after emails are found
• "Monitor more posts" → re-run with additional post URLs and deduplicate against this table

Tips

• Multiple posts = richer list. Scrape 3-5 recent posts from the same account to build a larger pool. Engagers across multiple posts are highly engaged — flag them.
• Repeat engagers are warmer leads. If someone engaged on 2+ posts, note that in the data — they're more likely to respond to outreach.
• Filter aggressively. Not all engagers are prospects. Use segment filtering to focus on decision makers and skip students, recruiters, etc.
• Respect rate limits. LinkedIn scraping providers have varying rate limits. Don't hammer the API — space out requests if scraping many posts.

Output