agentsclimarketplace
Sign in

Fb hook extractor

Skill sergebulaev/facebook-skills/.codex-marketplace/facebook-skills/skills/fb-hook-extractor

Reverse-engineer the hook from a high-share Facebook Page post, from its URL or pasted text. Identifies which of the 10 canonical 2026 Facebook formulas it uses (one-line opinion, tiny number, ask-the-page question, this-or-that, behind-the-scenes, story post, announcement, spotlight), explains why it earned shares or comments, and returns a blank template mapped to your topic. Use to learn from a Page post you admire. Not for writing your own (use fb-post-writer).From its SKILL.md

Install
npx -y skills add sergebulaev/facebook-skills --skill fb-hook-extractor

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 19 stars19 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

4.0 KB, 924 tokens by cl100k_base, as published. Nobody here has run it

Facebook Page Hook Extractor

Paste a high-share Page post URL (or the text). Get back: which hook formula it uses, the exact structure, why it worked, and a blank template you can fill with your own voice.

When to use

  • User finds a high-share Page post they want to study
  • User wants to replicate a specific Page's pattern
  • Before fb-post-writer, to seed a draft with a proven shape

Input

A Facebook Page post URL (any of the many shapes: /posts/, permalink.php, /watch/, share/p/). For a share/p/ link the page and post ids are hidden behind an opaque token, so paste the post text. The bundle has no built-in Page reader, so the user pastes the post text either way.

Output

  • Formula identified (FB1-FB10 from ../../references/hook-formulas.md) with a confidence score
  • Container: short post vs story post, and why that container fit the idea
  • Structural breakdown:
    • The hook line (and for a story post, how the first line earns the "See more" tap)
    • Body architecture (the beats, for a longer post)
    • The close (what earns the share or the comment)
    • Interaction-triggering devices (numbers, named entities, the question, the turn)
  • Primary goal the original chased (shares / comments / reactions)
  • Why it worked psychologically and algorithmically (meaningful interactions)
  • Blank template with {slot} markers matched to the original, ready for the user's topic
  • Cautions: anything in the original that would fail a 2026 audit (em dashes, AI vocab, "We are thrilled to announce", engagement bait, 5+ hashtags)

Steps

  1. Parse the URL. lib.url_parser.parse_facebook_url(url) returns page, post_id, share_token, url_type. If url_type is "share", tell the user the ids are hidden and ask for the post text.
  2. Get the text. This bundle has no built-in Page reader, so ask the user to paste the post (and any first comments, if relevant). If they later wire an Apify Page actor, read it automatically.
  3. Detect the container. A short post (one line / under ~80 chars) or a longer story post.
  4. Classify against the 10 formulas using features:
    • Short post: a flat opinion (FB1)? one hard number (FB2)? an easy question (FB3)? a binary this-or-that (FB4)? a relatable shared moment (FB5)? a behind-the-scenes glimpse (FB6)? a usable tip (FB7)?
    • Story post: a narrative with a turn (FB8)? a launch led human (FB9)? a named customer/team spotlight (FB10)?
  5. Score confidence. If two formulas fit, return the top 2 with fit scores.
  6. Extract structure. Label each part by its role. For a story post, map the first line (the "See more" tap), the beats, and the closer.
  7. Name the primary goal the original optimized for (shares / comments / reactions).
  8. Generate a blank template with {slot} markers matched to the original shape and the user's topic.
  9. Audit the source. Flag any AI tells in the original so the user does not copy them.

Example

See references/examples.md for worked teardowns.

Formulas reference

See ../../references/hook-formulas.md for the 10 canonical Facebook formulas with full skeletons and goal tags.

Files

  • SKILL.md - this file
  • references/classification-rules.md - feature extraction + scoring heuristics
  • references/examples.md - worked teardowns (short post and story post)

Related skills

  • fb-post-writer - use the extracted template to draft your own
  • fb-humanizer - audit your draft before shipping
  • fb-content-planner - study competitors' Pages while planning

What ships with it: 2 files

6.1 KB alongside SKILL.md

references/

Gives 0 of the 12 instructions most docs writing skills give in 924 tokens

Counted across 1,951 of the 3,904 authors here whose files we hold, read 2026-09-06

  • Use third-person for skill descriptionsin 54 of 1951, across 35 files
  • Start descriptions with Use whenin 43 of 1951, across 29 files
  • Run baseline scenarios before writing any skillin 40 of 1951, across 26 files
  • Use active voicein 40 of 1951, across 36 files
  • Map file responsibilities before defining tasksin 36 of 1951, across 29 files
  • Use checkbox syntax for tracking stepsin 35 of 1951, across 27 files
  • Ask one question at a timein 35 of 1951
  • Offer execution options after saving the planin 33 of 1951, across 24 files
  • Include complete code in every stepin 33 of 1951, across 27 files
  • Design units with clear boundaries and interfacesin 31 of 1951, across 23 files
  • Announce the skill usage at the startin 30 of 1951
  • Verify agent compliance after adding the skillin 29 of 1951, across 17 files

Said here and by no other author read

  • parse the provided facebook post url
  • request post text if url contains hidden identifiers
  • classify post against ten canonical facebook formulas
  • calculate confidence scores for formula matches
  • label structural components by their role
  • identify the primary engagement goal

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.