new-term-radar
Daily radar for new AI terms: names of new models, products and repos, caught on X, Hugging Face, GitHub and sitemaps before keyword tools list them.
翻譯準備中
只要一鍵,即可將任何病毒式 TikTok 影片轉換為您的商品廣告。適用於 TikTok Shop 賣家與聯盟行銷者——無需拍攝、無需剪輯。免費試用。
AI 翻譯
Turn any viral TikTok video into an ad for your own product in one click. For TikTok Shop sellers and affiliates — no filming, no editing. Free to try.
行銷GitHub @Clipcat-ai上線中營運者未確認
試用過了嗎? 點一下就好,不需要登入。
尚未偵測到更新。GitHub 發布版本後會顯示在這裡。
個人 GitHub 帳號下公開儲存庫的擁有者可以登入並驗證所有權,自行管理介紹,無需管理員核准。
登入並產生驗證檔案,在 30 分鐘內將其提交至此儲存庫的預設分支。其他分支或分叉儲存庫中的檔案無效。然後在下方檢查檔案,即可無需管理員核准直接管理專案。請保留檔案以維持管理權限。
如需登入,將前往登入頁面。登入後請回到此處,再次點擊所選的驗證按鈕。
在專案資料夾中執行 /nomorevibe verify,即可驗證所有權並自行管理介紹。
一分鐘完成驗證TikTok 電商影片生成、複製與分析 SKILL
AI 翻譯
TikTok e-commerce video generation, replication, and analysis SKILL
取自儲存庫 README
正在載入完整 README。 以下為已儲存的摘錄。
Clipcat Skill — AI Selling Videos for TikTok Shop
Turn any AI agent into a TikTok Shop video producer. Find viral TikTok videos, research TikTok Shop products, shops, creators and live rooms, break down why a video sells, generate a ready-to-shoot viral selling prompt from the library of real high-GMV AI selling videos, replicate a winning video with your own product, turn product photos into AI selling / UGC / talking-head videos, generate e-commerce images from a text prompt, publish the finished video to your own TikTok account, and download TikTok or Douyin videos — all from one CLI.檢視完整 README ↗Clipcat is the first Agent Skill dedicated to replicating winning ecommerce videos. It closes the loop from discovering proven formats → creating new winners through replication or product-to-video → publishing → analyzing performance and iterating. Turn Codex, WorkBuddy, Claude, OpenClaw, and other agents into capable TikTok Shop operators — and grow your TikTok business with far less manual work.
Four core capabilities — from finding proven inspiration to creating, publishing, and iterating on winning videos.
You need an agent such as Codex, WorkBuddy, or Claude, plus a Clipcat account and API key.
Send the appropriate command to your agent. Use the command that matches your computer’s operating system.
macOS / Linux / Git Bash:
curl -fsSL https://clipcat.ai/cli | bash
Windows PowerShell:
irm https://clipcat.ai/cli.ps1 | iex
Get your API key in Clipcat, then send this command to your agent:
clipcat config --api-key YOUR_API_KEY --base-url https://clipcat.ai
Replace YOUR_API_KEY with your own key, then send the complete command to your agent.
All five examples come from the Clipcat × Codex tutorial. Each follows the same structure: real conversation → playable result → prompt and notes.
Found a strong reference? Give your agent the TikTok link and product photos. Clipcat reconstructs the script and shot logic around your product.
The left player shows the proven source video. The right player shows the new product version created by Clipcat while preserving the original hook, shot rhythm, and selling structure. Click either player to watch it in place.
| Original video — proven reference | Replicated video — new product version |
|---|---|
| https://github.com/user-attachments/assets/a7f8f81c-8a0b-4184-9c4c-e4168bcf1d09 | https://github.com/user-attachments/assets/49015c45-9e5d-4177-b73c-6b2e3a121f85 |
The same workflow transfers across products and creative styles: Clipcat first identifies why the source works, then preserves its core structure while generating a new product-specific version.
| Original video — proven reference | Replicated video — same structure, new content |
|---|---|
| https://github.com/user-attachments/assets/500b630d-e524-4311-81e2-1567c20aaac5 | https://github.com/user-attachments/assets/a8c1bb64-04ea-44a6-af0b-b8b1d4d7f54f |
Use the Clipcat Skill to create a product-specific version of this proven reference.
TikTok reference: {COMPLETE_VIDEO_URL}
My product photos: {PRODUCT_IMAGES_OR_PATHS}
Keep the original hook, shot order, and pacing, but replace the product with mine.
Adapt any existing voiceover to my product. If there is none, do not add one. No subtitles.
Duration: {SECONDS}; model: {MODEL_ID}; voiceover language: {LANGUAGE}.
Show the full prompt, settings, and credit cost. Wait for my approval.
Notes
Preserve the proven hook and shot logic while adapting the product, claims, and voiceover. Read the full tutorial →
Upload product photos and a brief to create native-feeling demos, talking-head explainers, OOTD clips, reviews, and other TikTok Shop UGC.
| Generated result — click to play | ||
|---|---|---|
| https://github.com/user-attachments/assets/d90b1d54-6168-404e-908d-699420dd0563 |
Use the Clipcat Skill to create a selling video from my product photos.
Product photos: {PRODUCT_IMAGES_OR_PATHS}
Creative brief: {PRESENTER, SETTING, DEMO ACTIONS, AND SELLING_POINTS}
Market: {MARKET}; duration: {SECONDS}; aspect ratio: {9:16 OR 16:9}.
Model: {MODEL_ID}; voiceover language: {LANGUAGE}.
Use only the claims I provide. Show the prompt and credit cost before generating.
Notes
Clear product facts, presenter direction, setting, demo actions, and selling points lead to more reliable outputs. Read the full tutorial →
Connect discovery, selection, structural analysis, and replication in one workflow.
| Replicated after automatically finding a winning reference video | ||
|---|---|---|
| https://github.com/user-attachments/assets/a5f6c5be-1ee1-4627-95ed-93028f866e5e |
Use the Clipcat Skill to find popular selling videos for {PRODUCT_CATEGORY} in {TARGET_MARKET} from the last 7 days.
My product photos: {PRODUCT_IMAGES_OR_PATHS}
Explain query costs first. Show sources and available metrics, then select the best-fit reference.
Use its hook and pacing; adapt the product and voiceover to mine. No subtitles.
Duration: {SECONDS}; model: {MODEL_ID}; voiceover language: {LANGUAGE}.
Show the reference URL, full prompt, and generation cost. Wait for approval.
Notes
Keep the sources and available metrics. Choose a relevant reference without treating historical performance as a promise. Read the full tutorial →
Use a saved character reference or accessible portrait so a series can retain the same on-screen identity.
| Shoppable video with a consistent on-screen character | ||
|---|---|---|
| https://github.com/user-attachments/assets/28b3f3a0-8f55-42ab-96d5-0e3a724b3cd3 |
Use the Clipcat Skill to make a product video with my saved character.
Character: {AVAILABLE_CHARACTER_ID_OR_CHARACTER_IMAGE_URL}
Product photos: {PRODUCT_IMAGES_OR_PATHS}; TikTok reference: {COMPLETE_VIDEO_URL}
Keep the reference structure and pacing; adapt the product and voiceover to mine.
If the reference has an on-screen presenter, use my character reference. No subtitles.
Duration: {SECONDS}; model: {MODEL_ID_WITH_CHARACTER_SUPPORT}; voiceover: {LANGUAGE}.
Check character availability, show the settings and cost, and wait for approval.
Notes
Character references guide series consistency; results still depend on the model and source assets. Read the full tutorial →
Set up the workflow once to discover references, deduplicate them, generate outputs, log results, and send new creatives for review.
| Automatically find daily winners and replicate them: your always-on creative engine | ||
|---|---|---|
| https://github.com/user-attachments/assets/a5e0bb94-60bf-4cbc-bb00-f11efb237f7b |
Set up a recurring Agent task with the Clipcat Skill to keep new selling creatives coming.
Time: {DAILY_TIME}; timezone: {IANA_TIMEZONE}; market: {TARGET_MARKET}.
Category: {KEYWORDS}; product photos: {PERSISTENT_ASSET_REFERENCES}.
Find relevant videos from the last 7 days and deduplicate them against saved reference history.
Choose a reference, replace the product, and adapt the voiceover to my verified claims.
No subtitles. Duration: {SECONDS}; model: {MODEL_ID}; voiceover: {LANGUAGE}; per-run limit: {BUDGET}.
Confirm schedule, timezone, assets, budget, and unattended-generation authorization first. Do not auto-publish.
Notes
Automation needs persistent assets, deduplication history, a budget, and prior authorization. Automated generation is not automatic publishing. Read the full tutorial →
We also provide free access to the world’s largest library of AI shoppable videos and prompts.
It collects high-selling TikTok Shop videos across every active market and major category. Prompts are free to search and copy, and any matching format can be replicated with one click. Every prompt is reverse-engineered from a real TikTok Shop selling video — not written from scratch — and reconstructed shot by shot. Each entry includes the original video, so you can verify the structure directly.
We have published part of the Clipcat video and prompt library as Markdown in our GitHub Prompt Library, making it easy to browse, search, and reuse. The complete library contains 30,000+ videos across 15 markets and 30 categories, with new winners added daily. Our goal is simple: help more people create high-performing shoppable videos from real, verifiable references.
After installing the Clipcat Skill, try:
Search Clipcat’s prompt library for winning videos that match my product {PRODUCT_NAME_OR_DESCRIPTION}, then generate an adapted prompt for my product.
New users receive starter credits to explore Clipcat’s full workflow. Continue using the product by purchasing additional credits when needed.
Video and image generation, plus video and product queries, consume credits according to the task. Clipcat gives you one-click access to leading video models, including Seedance 2.5, Seedance 2.0, MiniMax H3, Wan 3.0, Omni, Veo 3.1, and Grok Imagine 1.5.
Before a video task is created, your agent confirms the requirements and credit cost with you. TikTok account authorization is required only once; future publishing does not require repeated authorization, and Clipcat supports multi-account management and publishing.
If Clipcat has made finding winners and creating winning product videos easier, please give us a ★. A Star won’t pay the bills—but it will encourage us to keep open-sourcing more useful projects for TikTok and cross-border ecommerce, helping you reach 10× growth in global commerce faster. One click. Three seconds. A real signal.