A social media prompt library is a structured collection of AI prompt chains — organized by platform — that transforms a single questionnaire input into platform-native social content. Rather than writing one generic post and copy-pasting it across channels, the library dispatches your brand context and key message to platform-specific prompt chains that produce Twitter threads, LinkedIn thought leadership, Instagram carousel copy, and TikTok scripts — each formatted and voiced for where it will live.
Most social media operations are built around a single person typing the same idea four different ways into four different boxes. The result is content that is technically present on every platform and effective on none. The Social Media Prompt Library was built to break this pattern: one questionnaire, four platform chains, four genuinely platform-native outputs that share an argument without sharing a sentence.
Why Social Media Content Fails at Scale
The default workflow at most teams is: write one post, copy-paste it to four platforms, lightly edit, hit publish. The output reads as if a stranger translated it across four different rooms. Twitter gets a paragraph. LinkedIn gets the same paragraph with hashtags. Instagram gets the same paragraph with emojis. TikTok gets a script that reads like the LinkedIn post being narrated.
The structural problem is that each platform has its own grammar — its own rules about what gets reach, what gets engagement, and what gets demoted. A Twitter thread that opens slow gets buried. A LinkedIn post that opens with a question gets throttled. An Instagram caption that depends on the image fails screen readers and the algorithm. A TikTok script written for the page rather than the ear gets skipped in three seconds.
The Library treats this as a generation problem rather than an editing problem. One questionnaire feeds four prompt chains, each with the corresponding platform grammar baked into the chain itself. The outputs are not paraphrases of one another. They are different pieces of content that share a thesis.
"One post copy-pasted four times is four underperforming posts. Platform-native generation isn't an editing pass — it's a generation architecture."
Tommy SaundersFounder, Windfield Real Estate
Platform-Native Prompt Architecture
The library is structured as four chains, each tailored to its platform. Twitter uses a four-prompt chain because thread structure benefits from explicit hook, expansion, engagement, and hashtag stages. LinkedIn uses three because the long-form post needs authority framing, narrative arc, and CTA optimization. Instagram and TikTok use shorter chains tuned to their respective output formats.
Twitter/X Thread Generator
The Twitter chain produces a thread, not a tweet. Stage one generates the hook tweet — a pattern interrupt that creates pull toward the rest of the thread. Stage two expands into 5–7 follow-up tweets, each tuned to stand alone while creating a loop into the next. Stage three optimizes for engagement signals (a question, a list teaser, or a stat). Stage four produces the hashtag strategy.
Stage two of the chain produces tweets two through seven. Each one reads cleanly on its own — important because the algorithm samples tweets out of order — while creating a small open loop that pulls the reader to the next. This is what makes the difference between a thread that gets a 4% expansion rate and one that gets 18%.
LinkedIn Thought Leadership Engine
The LinkedIn chain is shorter but more constrained. Authority framing demands first-person voice, data-led structure, and concrete claims. The narrative arc moves from a stated observation through a single example to a generalizable insight. The CTA closes with a question that drives reply-rate without using a question opener.
Instagram Carousel + Caption
Instagram needs two artifacts: the carousel storyboard (10 frames of visual concept + on-card copy) and the standalone caption that works without the image. The chain enforces a hook in the first line of the caption, image-independent prose, and a trailing hashtag block — never inline hashtag spam.
Four platform-native posts.
Zero copy-paste.
TikTok Script Generator
TikTok is the most structurally different platform — output is a video script, not a post. The chain produces three artifacts: a 3-second hook designed to win the swipe, a beat-by-beat script under 60 seconds with explicit pacing notes, and on-screen captions calibrated for thumb-reading. The first three seconds are non-negotiable.
TikTok's For You algorithm samples retention over the first three seconds before deciding whether to push the video to a wider audience. A weak opening doesn't get a second chance. The chain enforces a pattern-interrupt hook — a contradicting claim, a number, or a specific name — within the first beat.
The Cross-Platform Coherence Score
The Coherence Score measures how consistently the brand's key message, proof points, and strategic framing appear across all four outputs. The library evaluates outputs against the original questionnaire to ensure the Twitter thread, LinkedIn post, Instagram caption, and TikTok script all advance the same argument — even though each uses platform-native structure and voice.
Cross-Platform Coherence Score
A score above 0.85 indicates strong strategic alignment across all four platform outputs. Library-generated content typically scores 0.85–0.92. Copy-pasted content scores 0.40–0.55 because the source post drifts in voice and structure as it gets reformatted across platforms.
The score is built by comparing each output against the same questionnaire — measuring message alignment, proof-point retention, audience-tone match, and CTA consistency. When it drops, the failure is traceable to a specific chain and a specific stage, which means the fix is one prompt adjustment rather than a manual rewrite of four posts.
What this produces, in practice, is a content cadence where every platform performs against its own grammar without anyone on the team writing four versions of the same idea. The questionnaire becomes the only meaningful input. The chains do the platform translation. The score keeps everything tied to the same strategic argument. The team gets back the four hours it used to spend reformatting one post.