Linkedin Hook Extractor

Analyze any viral LinkedIn post URL to identify its hook formula, structure, why it worked, and generate a blank template for your own writing.

Security Scan

The skill's requested actions and files are consistent with its stated purpose (analyzing public LinkedIn posts and emitting a template), but a few external-capability assumptions are underspecified and worth checking before use.

Purpose & Capability

Name/description (teardown of viral LinkedIn posts) matches the instructions and included classification rules. The skill is instruction-only and does not ask for unrelated credentials, binaries, or config paths.

Instruction Scope

Runtime instructions are narrowly focused on parsing a LinkedIn URL, fetching the post text, extracting structured features, scoring formulas, and producing a template. However, the doc presumes availability of components not included in the bundle (e.g., lib.url_parser.parse_linkedin_url and a 'HarvestAPI' fetch mechanism) and refers to an external formulas file in another skill. Those assumptions are not declared or installed here and could cause the agent to reach out to external services or prompt the user for pasted content.

Install Mechanism

No install spec and no code files — nothing will be downloaded or written to disk by the skill itself. This minimizes install-time risk.

Credentials

The skill declares no required environment variables or credentials, which is appropriate for analyzing public posts. That said, the instruction to 'HarvestAPI preferred' implies use of an external scraping/fetch service that may require credentials or send data externally; those details are not declared in the skill.

Persistence & Privilege

The skill does not request permanent presence (always: false) and is user-invocable. It does not ask to modify other skills or system configuration.

Assessment

This skill appears coherent for analyzing public LinkedIn posts, but two practical checks before installing or using it: (1) Ask what 'HarvestAPI' refers to and whether it would cause the agent to call an external service (and if so, whether that service receives your post URLs or content). If HarvestAPI requires a token, confirm where that token would be stored and whether the skill would request it. (2) Verify whether your agent runtime provides the referenced helper 'lib.url_parser.parse_linkedin_url' or the related 'linkedin-post-writer' formula file; if not, expect the skill to prompt you to paste the post text. If you prefer not to have the agent contact external fetch services, use the fallback (paste the post text) when prompted.

Paste a viral LinkedIn post URL. Get back: which hook formula it uses, the exact structure, why it worked, and a blank template mapped to your topic.

When to use

• User finds a viral post they want to study
• User wants to replicate a specific creator's pattern (Jake Ward, Lara Acosta, etc.)
• Before linkedin-post-writer to seed a draft with a proven structure

Input

A LinkedIn post URL (any type: activity, share, ugcPost).

Output

• Formula identified (F1-F10 from linkedin-post-writer/references/hook-formulas.md) with confidence score
• Structural breakdown:
  • Hook lines (first 210 chars)
  • Body architecture (sections + what each does)
  • Close pattern
  • Reaction-triggering devices (numbers, named entities, vulnerabilities)
• Why it worked psychologically
• Blank template filled with slot markers matched to the original, ready for the user's voice
• Cautions: anything in the original post that would fail 2026 audit (em dashes, AI vocab, outdated tactics)

Steps

1. Parse URL. lib.url_parser.parse_linkedin_url → post_urn.
2. Fetch post body. HarvestAPI preferred; fall back to asking user to paste text.
3. Classify. Match against the 10 formulas using features:
   • First 2 lines: anaphoric? question? confession? number-led?
   • Body: numbered list? dated receipts? ledger? teardown?
   • Close: mirror question? identity reframe? commitment?
4. Score confidence. If multiple formulas fit, return top 2 with fit scores.
5. Extract structure. Pull each logical section and label it by formula role.
6. Generate blank template. Replace specifics with {slot} markers that match the user's topic.
7. Audit the source. Flag any AI tells in the original so the user doesn't copy them.

Example

Input: https://www.linkedin.com/posts/dharmesh_every-b2b-software-company-is-or-should-activity-7448808898326654978-iW20

Output:

• Formula: F10 Contrarian + Historical Receipts (confidence 0.72). Secondary: F5 Self-Proving Meta (0.28).
• Hook (first 210 chars): "Every B2B software company is (or should be) building an agentic version of their product."
• Body: single bold claim → 3 paragraphs of reasoning → specific list of product changes required
• Close: implicit call to action ("Seen this play out in your market yet?")
• Blank template:

  Every {category} {bold claim}.
  
  {Reasoning paragraph 1 — the forcing function}
  {Reasoning paragraph 2 — what it requires}
  {Reasoning paragraph 3 — what breaks if you don't}
  
  {Closing question that invites reader to take a side}
  

• Cautions: none (post is clean)

Formulas reference

See linkedin-post-writer/references/hook-formulas.md for the 10 canonical formulas with full skeletons.

Files

• SKILL.md — this file
• references/classification-rules.md — feature extraction + scoring heuristics

Related skills

• linkedin-post-writer — use the extracted template to draft your own
• linkedin-post-audit — audit your draft before shipping