Social caption writer
The open Skill Me catalog — every hosted skill as a portable, MIT-licensed SKILL.md
npx -y skills add SkillMedev/skills --skill social-caption-writerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 4 stars4 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.
What its author says it does
Copied from the file, not written here
Write one platform-native caption from a topic and brand voice, with a front-loaded hook, native length, and a single earned CTA. Use when the user gives a topic (and optionally brand voice or a described image/video) and asks for an Instagram, LinkedIn, X/Twitter, or TikTok caption. Do NOT use when the user wants a dedicated standalone LinkedIn post - use linkedin-post-writer instead; or a multi-post X thread - use tweet-thread-builder instead; or to adapt one existing piece of content into versions for several channels - use cross-platform-reformatter instead.
SKILL.md
3.6 KB, 700 tokens by cl100k_base, as published. Nobody here has run it
Social Caption Writer
Turn a topic plus brand voice into one platform-native caption that wins the first line and earns one specific action.
Workflow
- Confirm the platform (Instagram, LinkedIn, X/Twitter, or TikTok). If unstated, ask - every length, shape, and CTA below is platform-specific.
- Pin the brand voice. Use the traits the user gave. If they gave none, ask for 2-3 adjectives and one phrase the brand would never say before writing.
- Check the paired asset. If an image or video is described, write the caption to complement it (don't restate what's visible). If undescribed, ask - the caption's job depends on what it ships with.
- Reject thin or off-brand topics. If there's nothing to say, name that and propose a sharper angle instead of dressing up nothing.
- Write the hook into the first sentence - before the truncation cutoff (~125 chars on Instagram, ~140 on LinkedIn, the first line on X, one breath on TikTok). Lead with tension, a number, a contrarian claim, or a stakes-raising question. Never open with throat-clearing context.
- Fit native length and shape: Instagram 138-150 chars for punchy, or 1-3 tight paragraphs for storytelling; LinkedIn 1300-2000 chars with one-sentence lines and white space between them; X under 280, ideally ~100 so it breathes; TikTok short and curiosity-driven since the video carries the load. Use generous line breaks on Instagram and LinkedIn for mobile reading.
- Carry voice in word choice and rhythm, not adjectives: cut exclamation points for a dry/technical brand; let fragments and asides through for a playful one.
- End with exactly one specific CTA matched to platform behavior - saves/comments on Instagram, thoughtful replies on LinkedIn, reply/repost on X, follow/duet on TikTok. Make it concrete ("Save this for your next launch", not "engage with us"). Reference a link only when there's a real reason to leave the app.
Quality bar
- The first sentence stands alone as a reason to keep reading, with no context preamble before it.
- Length and line-break shape match the chosen platform's range above.
- The brand's voice is detectable in the verbs and rhythm, not just a sign-off.
- Exactly one CTA, specific and matched to how people act on that platform.
- Emojis used as accents, never as punctuation replacements.
Deliverable
Produce one ready-to-paste caption for the confirmed platform, formatted with its native line breaks, plus two support lines beneath it: the hook character count (proving it lands before the truncation cutoff) and any hashtags or link note called out separately so the user can drop or move them per their posting tool. One platform, one caption, one CTA - variants only if the user asks.
Do NOT
- Open with "In today's fast-paced world" or any throat-clearing context.
- Use engagement-bait: "comment YES below", tag-3-friends chains, follow-for-follow - platforms suppress it.
- Keyword-stuff the caption body.
- Stack multiple CTAs or default to "link in bio" with no real reason.
- Write a generic "social" caption that ignores the target platform's reading posture.
Gives 0 of the 12 instructions most docs writing skills give in 700 tokens
Counted across 1,637 of the 3,044 authors here whose files we hold, read 2026-08-06
- announce the skill at startin 54 of 1637, across 21 files
- convert legacy doc files before editingin 45 of 1637, across 7 files
- predict questions readers might askin 42 of 1637, across 3 files
- Generate clarifying questions for initial contextin 42 of 1637, across 3 files
- Create document scaffold with placeholder textin 42 of 1637, across 3 files
- Brainstorm content options for each sectionin 42 of 1637, across 3 files
- Test document with fresh context-less instancein 42 of 1637, across 3 files
- ask interview questions one at a timein 42 of 1637, across 26 files
- include exact file paths in every taskin 42 of 1637, across 15 files
- Apply surgical edits during refinementin 41 of 1637, across 2 files
- Offer structured workflow or freeformin 40 of 1637, across 1 file
- Ask for document meta-contextin 40 of 1637, across 1 file
Said here and by no other author read
- ask for brand voice traits if none are provided
- ask about the paired asset if undescribed
- reject off-brand topics and propose sharper angles
- fit the caption to native platform length
- carry brand voice in word choice and rhythm
- format the caption with native line breaks
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.