← ClaudeAtlas

caption-writerlisted

Use to write platform-native social media captions (post copy) in the brand's voice — the words that accompany a post on Instagram, LinkedIn, TikTok, Facebook, X/Twitter, Threads, Pinterest, YouTube, or Bluesky. Run when the user says "write a caption," "caption this," "post copy," "write the copy for this photo/video/Reel," "IG caption," "what should I write for this post," or needs the words to go with a visual. Reads brand-profile and voice first so captions sound like the brand, not generic AI. For long-form LinkedIn thought-leadership posts use linkedin-post-writer; for X/Threads multi-post threads use thread-writer; for video scripts use the video skills. This skill writes the caption itself.
social-media-skills/skills · ★ 5 · AI & Automation · score 73
Install: claude install-skill social-media-skills/skills
# Caption Writer A caption is not a description of the post. It's the part of the post that does the *work* — it stops the scroll, earns attention, and drives one action. Done well it carries the post; done generically it sinks even great visuals under emoji and hashtag noise. Two principles run through everything here: 1. **The first line is the whole game.** On every platform the caption is truncated — the reader sees one line and a "…more." If that line doesn't earn the tap, the rest is never read. Write the first line like its only job is to win the second line. 2. **Every caption serves one goal.** Awareness, comments, saves, clicks, community, or conversion — the goal dictates the structure and the call to action. A caption "to get engagement" is a caption with no goal. ## Step 0 — Read the foundation first (required) Load `brand-profile.md` and `voice.md` before writing a word. The voice fingerprint, the "never" list, and the brand guardrails are non-negotiable. A caption that doesn't sound like them is a failure even if it's clever. If neither exists, run `brand-profile` (and `voice-builder` if samples exist) first. ## Step 1 — Establish four inputs Don't write until you know these. Infer what you can; ask only what you can't: - **Platform** — the mechanics are different on each (see `references/platforms.md`). - **The asset** — what's in the photo/video/carousel? The caption complements the visual; it shouldn't narrate what the eye already sees.