tiktok-photo-mode — Claude Code skill: what it does, how to install
Plans native TikTok photo carousels and prepares WoopSocial publishing settings. Use it when a post should work as Photo Mode, not a video slideshow.
npx skills add social-media-skills/skills --skill tiktok-photo-mode -a claude-codeWhat tiktok-photo-mode does
This skill starts with brand-profile and goals-and-kpis, then maps the post into a swipe sequence. It creates the opening hook, staged reveals, cliffhangers, a final payoff, and one call to action. The structure targets completion rate rather than making reach claims.
The agent produces the slide outline, overlay wording, caption, and publish configuration. That configuration covers postType=PHOTO, privacy, comments, music, hashtags, search terms in overlays, and truthful disclosure flags. WoopSocial publishes the result as a native PHOTO post instead of an MP4 slideshow.
Images remain outside WoopSocial: a person or an image skill must create them. The layout uses 9:16 and 1080×1920, with content kept in the safe zone and usually 5–7 slides. Existing posts cannot be updated; while the status is NOT_STARTED, an edit means deleting and recreating the post.
What it helps with
- plan a TikTok photo carousel
- turn a topic into a swipe-by-swipe reveal
- write overlay copy and the caption
- prepare a WoopSocial PHOTO publish config
- check disclosure flags before publishing
Example prompts for the agent
- Plan a TikTok Photo Mode carousel using
brand-profileand the goal ingoals-and-kpis. - Create a
9:16slide plan with5–7slides, one CTA, a caption question, and WoopSocial settings forpostType=PHOTO. - Validate the publish configuration, hashtags, and
isYourBrandandisBrandedContentflags before posting.
How to install tiktok-photo-mode
1Skills CLI (Vercel)
npx skills add social-media-skills/skills --skill tiktok-photo-mode -a claude-code2Manually: put the files in place
~/.claude/skills/tiktok-photo-mode/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/tiktok-photo-mode/
1Skills CLI (Vercel)
npx skills add social-media-skills/skills --skill tiktok-photo-mode -a codex2Manually: put the files in place
.agents/skills/tiktok-photo-mode/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add social-media-skills/skills --skill tiktok-photo-mode -a cursor2Manually: put the files in place
.cursor/skills/tiktok-photo-mode/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install social-media-skills/skills tiktok-photo-mode2Manually: put the files in place
.github/skills/tiktok-photo-mode/SKILL.md1Skills CLI (Vercel)
npx skills add social-media-skills/skills --skill tiktok-photo-mode -a gemini-cli2Platform CLI
gemini skills install https://github.com/social-media-skills/skills --path skills/tiktok-photo-modeManually: put the files in place
.devin/skills/tiktok-photo-mode/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add social-media-skills/skills --skill tiktok-photo-mode -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about tiktok-photo-mode
How do I install tiktok-photo-mode in Claude Code?
Run npx skills add social-media-skills/skills --skill tiktok-photo-mode -a claude-code, then check ~/.claude/skills/tiktok-photo-mode/SKILL.md.
What does tiktok-photo-mode do?
It designs a TikTok-native photo carousel, writes the slide text and caption, and prepares WoopSocial settings for a PHOTO post.
Does the skill generate the carousel images?
No. A person or an external image skill creates the images; WoopSocial publishes the finished native PHOTO post.
Source text
SKILL.md · MIT
--- name: tiktok-photo-mode description: >- The TikTok Photo Mode format-execution skill. Use when someone wants a "TikTok photo carousel," "photo mode," "TikTok slideshow," "swipeable TikTok post," or to publish a multi-image TikTok post via WoopSocial (postType PHOTO). Builds a TikTok-native carousel that wins on completion rate — hook slide 1, progressive reveal with cliffhangers, reward + one CTA — AND sets the exact WoopSocial PHOTO publish config (privacy, comments, music, truthful disclosure flags). Uses the SWIPE framework. Reads brand-profile + goals-and-kpis first. The agent writes the slide plan + overlay copy + caption + config; the IMAGES are made by a human/image skill, not WoopSocial; WoopSocial publishes the native PHOTO post; metrics never fabricated and photo mode is never promised as a reach hack. Distinct from tiktok-script (VIDEO) and carousel-writer (cross-platform copy). version: 1.0.0 --- # tiktok-photo-mode A **format-execution skill** — produce a TikTok-native Photo Mode carousel (craft for completion) **and** the exact WoopSocial PHOTO publish config. The agent writes the **plan + overlay copy + caption + config**; the **images are made by a human or an image skill**; **WoopSocial publishes the native PHOTO post**. ## The POV: win completion, don't chase a reach hack TikTok treats Photo Mode **like video** and ranks it on **completion rate** — what % swipe to slide 2+ and through the end. So every choice serves "make them swipe to the end": a slide-1 hook that promises a payoff, one idea per slide with cliffhangers, a reward + one CTA at the end. But be honest: photo mode is a **high-ceiling, niche-dependent format**, still outnumbered ~5:1 by video with **declining average per-post performance** as it saturates — it can punch far above its weight for the right content type, and underperforms video when forced. **Never promise virality.** Design **9:16 (1080×1920)**, not Instagram's square, and publish it as a **native PHOTO post** (not an MP4 slideshow). ## Read these first 1. **brand-profile** — niche + voice (does a swipeable format fit?). 2. **goals-and-kpis** — the goal (saves / comments / follows / reach). ## The framework: SWIPE (Depth: `references/the-swipe-framework.md`.) - **S — Stop the scroll on slide 1:** the thumbnail hook (bold visual + overlay question/claim) that promises a payoff. - **W — Win the niche fit:** pick a swipeable type (listicle / before-after / tutorial / hot take / photo dump / story / myth-bust); don't force it — photo mode is niche-dependent. - **I — Itemize for completion:** one idea per slide, progressive reveal, cliffhangers; 5–7 slides (max 35); 9:16 1080×1920, content centred in the safe zone. - **P — Pay off + prompt:** reward on the final slide + one CTA; question in the caption; don't beg for likes. - **E — Equip the publish config:** `postType=PHOTO`, `privacyLevel` (from platform-inputs), `allowComment`, `autoAddMusic` (applies to PHOTO), truthful `isYourBrand`/`isBrandedContent`, `DIRECT_POST` vs `MEDIA_UPLOAD`; 1–2 hashtags + overlay keywords (TikTok SEO); **validate before publish.** ## The reality (verify-quarterly) The honest performance tension (high-ceiling vs video-dominance + declining averages); completion rate as the primary signal; 9:16 1080×1920 + safe zone; 5–7 slides; hook / progressive reveal / cliffhanger / reward; sound (autoAddMusic) + text overlays as TikTok SEO; 1–2 hashtags; native-vs-MP4; winning content types + engagement tactics: `references/photo-mode-2026-reality.md`. Content-type menu, slide blueprint, the full PHOTO publish-config block + worked examples: `references/carousel-recipes-and-publish.md`. ## Honest scope (never violate) - **The agent writes the slide plan + overlay copy + caption + publish config.** The **images are produced by a human or an image skill** (`image-prompt`/`ideogram`/`nano-banana`) — **WoopSocial doesn't generate images.** - **WoopSocial publishes the native PHOTO post** (`postType=PHOTO`, raw-bytes upload) and **auto-discloses AI on TikTok.** **No update endpoint** → edit = delete + recreate (only while `NOT_STARTED`). - **Disclosure flags set truthfully** (`isBrandedContent` when paid/gifted; `isYourBrand` for own-brand promo) — never falsely "organic." **Never promise virality / a "5×"; never fabricate metrics** (native analytics only). (Scope, distinctions + connections: `references/scope-and-connections.md`.) ## Distinct from its siblings (route correctly) **tiktok-photo-mode (this)** = the TikTok-native PHOTO carousel (completion craft + the WoopSocial PHOTO publish config) · **tiktok-script** = the TikTok VIDEO script (duet/stitch/cover live there) · **carousel-writer** = cross-platform carousel copy craft · **reels-script**/**youtube-shorts** = video · **caption-writer**/**hook-writer** = the caption + overlay lines this skill arranges. ## Where this connects Reads first: **brand-profile**, **goals-and-kpis**. Copy: **hook-writer**, **caption-writer**. Images (external to WoopSocial): **image-prompt**/**ideogram**/**nano-banana**, **carousel-writer** (adapt down to 9:16). Publish: **scheduling-and-queue → WoopSocial** (native PHOTO), **platform-specs-and-validation** (validate + postType/field rules), **content-calendar**. Reuse/measure: **content-recycling** (→ IG/ Pinterest), **analytics-and-reporting** (completion/saves), **experimentation-and-ab-testing** (A/B the first slide). Images stay external. ## Definition of done A swipeable content type matched to the niche and goal; a slide-1 hook that promises a payoff; a 5–7 slide progressive reveal with cliffhangers, one idea per slide, designed 9:16 1080×1920 with content in the safe zone; a final-slide reward + one CTA and a caption question (not a like-beg); 1–2 hashtags + overlay keywords for TikTok SEO; the complete WoopSocial PHOTO publish config (postType PHOTO, privacyLevel, allowComment, autoAddMusic, truthful isYourBrand/isBrandedContent, postMode) validated before publish; images produced externally, WoopSocial publishing the native PHOTO post with AI auto-disclosure, edits via delete+recreate, nothing fabricated and no promised virality; correctly distinguished from tiktok-script and carousel-writer.
© social-media-skills · MIT · social-media-skills/skills