Skip to content
Back to search

arcads-claude-code

Arcads external API: agent skills, prompting library, and Cursor/Claude workspace

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/krusemediallc--arcads-claude-code/badge.svg)](/listings/krusemediallc--arcads-claude-code)

HTML

<a href="/listings/krusemediallc--arcads-claude-code"><img src="/api/listings/krusemediallc--arcads-claude-code/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

./scripts/setup.sh

This will: - Sign up if you need an Arcads account: [arcads.ai/?via=claude-code](https://arcads.ai/?via=claude-code) - Ask for your **Arcads API key** (find it at [app.arcads.ai/settings/api](https://app.arcads.ai/settings/api)) - Save it securely in `.env` (never committed to git) - Verify your connection to Arcads - Create your personal `MASTER_CONTEXT.md` workspace file

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 399 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # Arcads AI Video — Agent Skill Pack
  2. Create AI marketing videos and images using your [Arcads](https://arcads.ai/?via=claude-code) account, powered by AI agents in **Claude Code** or **Cursor**. Supports the full Arcads creative stack — **Seedance 2.0** (flagship video), **Sora 2**, **Veo 3.1**, **Kling 3.0**, **Grok Video**, **Nano Banana 2 / Pro / Edit**, **ChatGPT Image 2**, **OmniHuman**, and **Audio-driven** — plus a 37-template static Meta image-ad library and a pipeline for **Pixar-style** and **claymation** animated ads.
  3. ## Anatomy of a great AI song ad
  4. A visual guide to the nine story beats of a narrative AI song ad, why each section works, and the music, visuals, and editing choices that support the story.
  5. [![Anatomy of a great AI song ad: nine illustrated story beats from the opening hook to the final invitation, with an explanation of each section and supporting music, picture, and editing principles.](docs/images/ai-song-ad-anatomy.png)](docs/images/ai-song-ad-anatomy.png)
  6. [View the full-size infographic](docs/images/ai-song-ad-anatomy.png)
  7. ## 🎥 Watch the walkthrough
  8. [![How To Build Entire AI Ad Campaigns With Claude Code (free repo)](https://i.ytimg.com/vi/HHGQN9Zqaxo/hqdefault.jpg)](https://youtu.be/HHGQN9Zqaxo)
  9. **[How To Build Entire AI Ad Campaigns With Claude Code (free repo)](https://youtu.be/HHGQN9Zqaxo)** — full build tutorial by Mr. Paid Social.
  10. ## Level up your media buying with AI
  11. This repo handles the **creative** side. If you want the full system — scaling, ROAS, automation, and the AI workflows behind 8-figure ad accounts — come build with me inside **[The AI Ad Alchemists](https://skool.com/mrpaidsocial)**.
  12. It's a private Skool community of **460+ media buyers** managing 8-figure ad accounts, run by me (Caleb — aka "Mr. Paid Social") drawing on 12 years and **$150M+ in ad spend**. Inside you get:
  13. - **Meta Masterclass** — the exact systems I use to scale Meta ads (valued at $1.2k)
  14. - **Custom GPTs** for ad copywriting and compliance
  15. - **Ad swipe files** + breakdowns of what's working *right now*
  16. - **Airtable & Google Sheets scaling systems** — the operational backbone behind 8-fig accounts
  17. - **AI tool walkthroughs** — including the Ad Agent (built in Claude Code) and the GenAI system in Airtable I'm shipping next
  18. - **Monthly group calls** + guest speakers
  19. - **Direct access** to me and the network
  20. If you're using this repo to crank out creative, the community is where you learn to **turn that creative into ROAS at scale**.
  21. **→ [Join The AI Ad Alchemists — $97/month](https://skool.com/mrpaidsocial)**
  22. ## Facebook Ad Library → Airtable swipe file
  23. Build a searchable library of competitor image ads with the included [Meta Ad Library → Airtable workflow](scripts/meta-ad-library-import/README.md). Supply up to **10 advertiser or brand entries**; the default collection limit is **100 active image ads per entry**, subject to what Meta's API returns.
  24. The workflow collects ad metadata through **Meta's Ad Library API**, then uses authorized automated headless Chrome to retrieve images from the API's snapshot links. An agent or person reviews each downloaded creative before it is imported into Airtable with:
  25. - **Image and thumbnail attachments**, plus the public Ad Library link.
  26. - **Brand and advertiser details**, ad copy, dates, platforms and verified CTA links.
  27. - **Relevant creative tags and format choices** matched to your table's schema.
  28. - **An individual recreation prompt** describing the image's layout, typography, colors, subject and text hierarchy for your chosen image generator, such as ChatGPT Image 2.5.
  29. Ad-ID checks skip previously imported ads, and the importer verifies that Airtable has received the attachments. The advertised brand stays separate from the Facebook Page, so a shared advertiser such as Skool can be filtered into the right communities.
  30. **Try this with your agent:**
  31. > "Follow scripts/meta-ad-library-import/README.md. Import up to 100 active static image ads for each of these Facebook Page IDs: [IDs], into [Airtable URL]. Use the credentials in my .env. I authorize automated headless extraction of the API-provided snapshots. Inspect every image, fill all applicable fields, select relevant existing tags and formats, and write an individual ChatGPT Image 2.5 recreation prompt. Check for duplicate ad IDs and verify the uploaded attachments."
  32. You need a Meta token with Ad Library access, an Airtable PAT or connected Airtable MCP, and Chrome with the extraction dependencies. API coverage varies by delivery country and access; the tested setup uses explicit UK/EU countries. This workflow writes prompts and does not generate replacement ads.
  33. **[Full setup and commands](scripts/meta-ad-library-import/README.md)** · **[Agent handoff guide](scripts/meta-ad-library-import/README.md#using-this-workflow-with-an-ai-agent)** · **[Advertiser example](scripts/meta-ad-library-import/examples/advertisers.json)**
  34. ## Prerequisites
  35. The agent and the basic Arcads workflows (image generation, video generation, polling) work with just **Python 3.10+** and the API key from setup. Some multi-step pipelines need a few extra CLI tools:
  36. | Tool | Required for | Install (macOS) |
  37. |---|---|---|
  38. | **Python 3.10+** | Everything (the image-ad generators are stdlib-only — no pip install) | preinstalled or `brew install python@3.12` |
  39. | **`ffmpeg`** | Pixar-style ad, claymation ad, caption-video (stitching + chroma-key overlay) | `brew install ffmpeg` |
  40. | **`jq`** | Several bash scripts (pixar-style-ad, etc.) | `brew install jq` |
  41. | **Node.js + `npx hyperframes`** | Caption burn-in workflow | `brew install node` (the skill runs `npx` on demand) |
  42. | **`whisper`** Python package | Caption transcription | `pip install openai-whisper` (or `pip3`) |
  43. | **`meta-ad-builder` deps** | Publishing to Meta Marketing API | `pip install -r shared/skills/meta-ad-builder/scripts/requirements.txt` |
  44. | **Chrome + Selenium/Pillow** | Meta Ad Library image extraction | Chrome, then install [workflow dependencies](scripts/meta-ad-library-import/README.md#2-extract-the-images) in a virtual environment |
  45. The image-ad generator scripts (`chatgpt-image-ad`, `nano-banana-image-ad`, `image-ad-clone`) are intentionally stdlib-only — no extra installs needed. The deps above are only required when you invoke the matching multi-step workflow.
  46. Linux users: `apt install ffmpeg jq nodejs python3`. Windows users: WSL2 recommended; the shell scripts assume bash.
  47. ## Get started (5 minutes)
  48. ### 1. Clone this repo
  49. ```bash
  50. git clone https://github.com/krusemediallc/arcads-claude-code.git
  51. cd arcads-claude-code
  52. ```
  53. ### 2. Run setup
  54. ```bash
  55. ./scripts/setup.sh
  56. ```
  57. This will:
  58. - Sign up if you need an Arcads account: [arcads.ai/?via=claude-code](https://arcads.ai/?via=claude-code)
  59. - Ask for your **Arcads API key** (find it at [app.arcads.ai/settings/api](https://app.arcads.ai/settings/api))
  60. - Save it securely in `.env` (never committed to git)
  61. - Verify your connection to Arcads
  62. - Create your personal `MASTER_CONTEXT.md` workspace file
  63. ### 3. Open in your AI editor
  64. **Claude Code:** Open the folder. A `SessionStart` hook runs and prints an orientation banner showing which skills are installed, whether your `.env` and `MASTER_CONTEXT.md` are set up, and where the docs live.
  65. **Cursor:** Open the folder. Same skills are exposed at `.cursor/skills/`.
  66. ### 4. Start creating
  67. The agent handles API calls, polling, prompt engineering, file organization, and cost confirmation. Workflows are grouped by what you want to make.
  68. ---
  69. ### 🎬 Seedance 2.0 videos (the flagship — start here)
  70. Seedance 2.0 is the most flexible model in the stack — 4–15s clips, native audio, image-to-video or video-to-video, reference images, multiple shot styles. Five prompt formulas ship with the skill:
  71. #### UGC selfie-style product review
  72. > "Make a 12-second Seedance UGC video — woman in a kitchen, holding the product, says she stopped buying [competitor]"
  73. Uses the 9-layer UGC formula tuned for Seedance 2.0 (iPhone-shot aesthetic, natural eye-contact breaks, casual delivery). See `skills/arcads-external-api/prompting/prompt-library/seedance-2-ugc.md`.
  74. #### Premium product reveal (no person)
  75. > "Premium reveal of [product] — dark void, text narrative, hero rotation"
  76. Dark-void aesthetic, text overlays narrating the product's positioning, no person on screen. See `skills/arcads-external-api/prompting/prompt-library/seedance-2-premium-reveal.md`.
  77. #### Product hero with elemental effects
  78. > "Seedance product hero — water splash, mist, slow rotation"
  79. Splash, mist, light rays, slow rotation. See `skills/arcads-external-api/prompting/prompt-library/seedance-2-product-hero.md`.
  80. #### Studio lookbook with voiceover
  81. > "Studio lookbook of [product] — multi-look, polished, with voiceover script"
  82. Polished editorial / lookbook style, multi-shot, with embedded dialogue. See `skills/arcads-external-api/prompting/prompt-library/seedance-2-studio-lookbook.md`.
  83. #### Feature walkthrough demo
  84. > "Seedance feature walkthrough — fast-paced, show off [features]"
  85. Fast-paced product-demo cuts. See `skills/arcads-external-api/prompting/prompt-library/seedance-2-feature-walkthrough.md`.
  86. ---
  87. ### 🎬 Other video models
  88. #### Sora 2 (text-to-video, up to 20s)
  89. > "Generate a 16-second Sora video of [scene]" — optionally drop a product photo as a style reference.
  90. Sora 2 handles longer durations than Veo. The agent auto-selects `duration` from your script's word count (~2.5 words/sec). Sora 2 remix is also supported via `POST /v1/sora2/remix/video` for remixing an existing asset.
  91. #### Veo 3.1 (starting-frame animation)
  92. > "Animate this Nano Banana still into an 8-second Veo with dialogue"
  93. Veo 3.1 with `startFrame` is the standard path for **UGC stills → video**. The video starts from your exact image with natural human motion and embedded dialogue. The agent confirms the dialogue separately before generating (the MANDATORY dialogue gate).
  94. #### Kling 3.0 (b-roll / scene)
  95. > "Make a 5-second b-roll clip of [scene]" or "Generate a scene of [environment]"
  96. Kling 3.0 is the b-roll / scene workhorse. Hits the dedicated `POST /v1/b-roll` and `POST /v1/scene` endpoints.
  97. #### Grok Video
  98. > "Generate a Grok video of [scene]"
  99. Hits `POST /v2/videos/generate` with `model: "grok-video"`.
  100. #### OmniHuman / Audio-driven
  101. > "OmniHuman avatar of [person] delivering [script]" or "Audio-driven video lip-synced to [audio file]"
  102. `POST /v1/omnihuman` for talking-avatar workflows; `POST /v1/audio-driven` for lip-sync against an audio file.
  103. ---
  104. ### 🖼️ Image generation (Nano Banana + ChatGPT Image 2)
  105. #### Create a new AI influencer (10-image character sheet)
  106. > "Create a new AI influencer — 22-year-old college student with freckles, golden-hour kitchen lighting"
  107. Two-pass workflow: (1) generate a hero front portrait via Nano Banana, get your approval, (2) generate 9 additional angles (3/4 views, profile, close-up, expressions) with the hero as the reference. All 10 saved to `references/influencers/` for future reuse.
  108. #### UGC product selfie still
  109. > "Generate a UGC selfie of Sofia holding [product] in her bedroom"
  110. Combines your character hero + product photo + style references from `references/aesthetics/ugc-selfie/` into an authentic-looking iPhone selfie frame grab. Includes skin realism and camera imperfections to fight AI's polished default.
  111. #### Product showcase still (Nano Banana → Veo / Seedance video)
  112. > "AI person holding [product] talking about [feature]"
  113. Two-step: Nano Banana still of person + product → user approves → start-frame → video via Veo 3.1 or Seedance 2.0.
  114. #### Recreate an influencer from a reference photo
  115. > "Recreate this influencer's look from this reference photo"
  116. Two-step: Nano Banana still from `refImageAsBase64` → user approves → Veo 3.1 with `startFrame`.
  117. #### Nano Banana model choice
  118. Default is `nano-banana-2`. Use `model: "nano-banana"` for **Nano Banana Pro** (Gemini 3 Pro Image — higher fidelity, locks character identity tighter across reference batches). `nano-banana-edit` for inpainting.
  119. ---
  120. ### 📸 Static Meta image ad creative (37-template library)
  121. > "Make me an Apple Notes-style ad for my product" / "Generate a Forbes editorial ad" / "Clone this comparison-table ad as a template"
  122. A four-skill family for static Meta image ads with a shared library of **37 validated prompt templates** (Apple Notes lists, editorial hero, fake Google search, comparison tables, sticky-note flatlays, fake Slack threads, ChatGPT-conversation ads, iMessage screenshots, magazine cover, billboard, museum exhibit, weather forecast UI, scratch-off ticket, founder letter, dating-app card, more).
  123. - **`chatgpt-image-ad`** — typography-heavy / UI-mimicry creatives (gpt-image-2)
  124. - **`nano-banana-image-ad`** — photoreal / lifestyle / multi-reference creatives (Nano Banana 2 / Pro / Edit)
  125. - **`image-ad-clone`** — single backend-agnostic skill that reverse-engineers any existing ad image into a new library entry (asks which generator to validate against at Phase 1; optionally cross-validates against the other at Phase 8)
  126. Output is image files. Pair with the `meta-ad-builder` skill to publish as paused Meta ads. **Read `shared/skills/image-ad-prompting/OVERVIEW.md` first** — it has the decision tree (which backend for which template), the aspect-ratio compatibility matrix per backend, and the standard generate / clone workflows. Live-validated end-to-end against the Arcads API.
  127. ---
  128. ### 🎞️ Multi-step animated ad pipelines
  129. #### Pixar-style 3D animated ad
  130. > "Make a Pixar-style ad for [product] — anthropomorphized mascot, 8-beat story arc"
  131. Lock cast sheet → ChatGPT Image 2 storyboard stills (sequential, prior frame as ref for identity lock, max 5 `referenceImages`) → Seedance 2.0 image-to-video per beat → ffmpeg stitch + burn captions. See `shared/skills/pixar-style-ad/prompting/guide.md`.
  132. #### Claymation / Aardman-style ad
  133. > "Make a claymation ad — sculpted plasticine characters, narrator-driven, 60–115s"
  134. Same backbone as Pixar with an 8-beat narrator-driven story arc and clay textures. ChatGPT Image 2 storyboard → Seedance 2.0 i2v → ffmpeg stitch with optional `fps=12,fps=24` stop-motion judder. VO generated externally (ElevenLabs) and mixed in post. See `shared/skills/claymation-ad/prompting/guide.md`.
  135. #### YouTube thumbnails (5 CTR formulas)
  136. > "Make 6 YouTube thumbnail variations with my face and product"
  137. Specialized `generate-youtube-thumbnail` skill: peace-sign/branding, real-vs-AI comparison, terminal flow, reaction shock, before/after split. Likeness lockdown via 5+ face references. Parallel batch firing against Nano Banana 2. See `skills/generate-youtube-thumbnail/`.
  138. #### Burn captions onto a finished video
  139. > "Add captions to this MP4"
  140. Out-of-band post-step (no Arcads call) that works on any source — Pixar, claymation, UGC, B-roll. HyperFrames + Whisper `medium.en` for transcription → group word-level transcript into reading phrases → render captions-only HTML over `#ff00ff` magenta → ffmpeg chroma-key overlay. See `shared/skills/caption-video/prompting/guide.md`.
  141. ---
  142. ### 🔄 Reverse-engineer existing creative
  143. #### Analyze a reference video into a Seedance template
  144. > "Reverse-engineer this video into a reusable Seedance template"
  145. The `analyze-video` workflow under `skills/arcads-external-api/prompting/analyze-video/` extracts the structure of a reference video into a parameterizable Seedance 2.0 prompt template.
  146. #### Clone an existing video ad for a different product
  147. > "Clone this video ad for our new product"
  148. `skills/arcads-external-api/prompting/clone-ad/` — end-to-end: analyze the reference → adapt to the new product → generate. The companion to `analyze-video` when you want to ship the cloned version directly.
  149. #### Clone a static image ad into the prompt library
  150. > "Reverse-engineer this image ad as a reusable template"
  151. The `image-ad-clone` skill produces parameterizable entries for the 37-template library (see above).
  152. ---
  153. ### 📤 Publish creatives as paused Meta ads
  154. > "Publish this approved creative as a paused Meta ad in my account"
  155. The cross-API `meta-ad-builder` skill (in `shared/skills/`) takes a finished creative path and uploads it via the Meta Marketing API. Every ad is created PAUSED — you review and launch manually in Ads Manager. Also has a research path to pull top-spending ads and competitor ads. Auth via `META_*` keys in `.env`.
  156. ## What's in the box
  157. | Path | What it does |
  158. |------|-------------|
  159. | `skills/arcads-external-api/` | **The core skill.** API reference, prompting guide, per-model prompt libraries (Seedance / Sora / Veo / Kling / Nano Banana), analyze-video + clone-ad sub-workflows. |
  160. | `skills/generate-youtube-thumbnail/` | 5 CTR-tested YouTube thumbnail formulas with parallel batch firing against Nano Banana 2. |
  161. | `skills/chatgpt-image-ad/` | Static Meta image-ad creatives via gpt-image-2 (typography / UI mimicry). Live-validated. |
  162. | `skills/nano-banana-image-ad/` | Static Meta image-ad creatives via Nano Banana 2 / Pro / Edit (photoreal / lifestyle). Live-validated. |
  163. | `skills/image-ad-clone/` | Reverse-engineer an existing ad image into a reusable library entry. Backend-agnostic — asks at Phase 1 whether to validate via gpt-image-2 or Nano Banana, optionally cross-validates against the other at Phase 8. |
  164. | `shared/skills/image-ad-prompting/` | Shared brain for the image-ad ecosystem: 37 validated templates, safety suffixes, entry format, `OVERVIEW.md`. |
  165. | `shared/skills/pixar-style-ad/` | Cross-API recipe: 8-beat anthropomorphized mascot ad via GPT Image 2 storyboard + Seedance 2.0 i2v. |
  166. | `shared/skills/claymation-ad/` | Cross-API recipe: Aardman-style 8-beat clay narrative ad; same backbone as Pixar with stop-motion judder option. |
  167. | `shared/skills/caption-video/` | Out-of-band post step: HyperFrames + Whisper + ffmpeg chroma-key to burn captions onto any finished MP4. |
  168. | `shared/skills/meta-ad-builder/` | Publish finished creatives as paused Meta ads via the Meta Marketing API. |
  169. | `shared/scripts/check-context.sh` | SessionStart banner — lists installed skills, checks `.env` / `MASTER_CONTEXT.md` status, surfaces ecosystem pointers. Hooked into `.claude/settings.json`. |
  170. | `MASTER_CONTEXT.template.md` | Template for your workspace context (credit costs, brand voice, learnings). |
  171. | `MASTER_CONTEXT.md` | Your personalized copy (created by setup, not committed to git). |
  172. | `.env` | Your API key (created by setup, never committed). |
  173. | `scripts/setup.sh` | One-time setup. |
  174. | `scripts/sync-skill.sh` | Copies skill edits to `.claude/` and `.cursor/` directories. |
  175. | `scripts/check-arcads-env.sh` | Tests API connectivity. |
  176. | `references/` | Drop reference images here (influencers, products, aesthetics) — gitignored. |
  177. | `logs/arcads-api.jsonl` | Per-call audit log: model, duration, resolution, reference counts, `creditsCharged`. Powers cost-estimation accuracy across sessions. |
  178. ## Your API key
  179. Your key authenticates with the Arcads API. During setup you paste it once and the agent uses it from `.env` automatically. You never need to paste it into chat.
  180. Need an Arcads account first? Create one here: **[https://arcads.ai/?via=claude-code](https://arcads.ai/?via=claude-code)**
  181. Find your key: **[Arcads Dashboard > Settings > API](https://app.arcads.ai/settings/api)**
  182. For Meta-ad publishing (the `meta-ad-builder` skill), you'll also need `META_ACCESS_TOKEN` and `META_AD_ACCOUNT_ID` in `.env` — the `.env.example` has placeholder rows.
  183. ## Project memory
  184. `MASTER_CONTEXT.md` is your workspace's living memory. The agent reads it at the start of every session and writes learnings back. It stores:
  185. - **Default product** — auto-populated on first use so you're never asked "which product?" again
  186. - **Default project / folder** — session output organized in the Arcads dashboard automatically
  187. - **Credit costs** — you fill in once (or the agent asks), then every session has them
  188. - **Image hosting** — where you stage reference images if a workflow needs hosted URLs
  189. - **Brand voice** — optional tone, audience, and word preferences
  190. - **API learnings** — universal Arcads quirks that help the agent work better
  191. - **Changelog** — dated notes from each session
  192. ## Supported models
  193. | Model | Type | Best for | Notes |
  194. |-------|------|----------|-------|
  195. | **Seedance 2.0** | Video (4–15s) | Flagship video model. UGC, premium reveal, product hero, lookbook, feature walkthrough. Native audio. | `model: "seedance-2.0"`. 5 prompt formulas ship. Mutually exclusive: `referenceVideos` vs `referenceImages`. |
  196. | **Sora 2** | Video (up to 20s) | Long-duration text-to-video, image-to-video with product photo as style ref. | `model: "sora2"`. Duration enum: `[4, 8, 12, 16, 20]`. Remix: `POST /v1/sora2/remix/video`. |
  197. | **Veo 3.1** | Video (~8s) | Animating a starting frame (UGC stills → video). Best for character / influencer flows. | `model: "veo31"`. `startFrame` and `referenceImages` are mutually exclusive — default `startFrame` for single person photos. |
  198. | **Kling 3.0** | Video (5s or 10s) | B-roll and scene generation. | Hits `POST /v1/b-roll` / `POST /v1/scene` directly. |
  199. | **Grok Video** | Video | Text-to-video via xAI's video model. | `model: "grok-video"`. |
  200. | **Nano Banana 2** | Image | Default still-image model. UGC stills, character sheets, product shots, influencer recreation, image-ad creatives. | `model: "nano-banana-2"`. |
  201. | **Nano Banana Pro** | Image | Premium image quality (Gemini 3 Pro Image). Locks character identity tighter across batches. | `model: "nano-banana"` (the bare string maps to Pro on Arcads). |
  202. | **Nano Banana Edit** | Image | Inpaint / edit an existing image. | `model: "nano-banana-edit"`. |
  203. | **ChatGPT Image 2** | Image | Typography-heavy / UI-mimicry static ad creatives. Used by `chatgpt-image-ad` skill + the Pixar / Claymation storyboard pipelines. | `model: "gpt-image-2"`. Max 5 `referenceImages`. Aspect ratios: `1:1`, `16:9`, `9:16`. |
  204. | **OmniHuman** | Video | Talking-avatar / lip-sync workflows. | `POST /v1/omnihuman`. |
  205. | **Audio-driven** | Video | Lip-sync a video to a supplied audio file. | `POST /v1/audio-driven`. |
  206. Cost is presented as an **estimate** before every generation; the agent reads `logs/arcads-api.jsonl` for historical `creditsCharged` values matching your config. Always confirm exact pricing in the Arcads dashboard if it matters for budgeting.
  207. ## Reference images
  208. Drop images into the `references/` folder and the agent will use them automatically:
  209. - **`references/influencers/`** — Photos of people to recreate as AI-generated content (and saved character sheets)
  210. - **`references/products/`** — Product photos for showcase videos and hero images
  211. - **`references/aesthetics/`** — Style references organized by vibe (`ugc-selfie/`, `cinematic/`, etc.)
  212. Images stay local — the folder contents are gitignored. The agent auto-upscales any reference below 1024px (the API's min-size floor) using Lanczos before submitting.
  213. ## Editing skills
  214. Each skill's canonical source lives in `skills/<name>/`. After editing any file there, run:
  215. ```bash
  216. ./scripts/sync-skill.sh
  217. ```
  218. This copies your changes to `.claude/skills/` and `.cursor/skills/` (which are gitignored — they're generated copies). The `SessionStart` hook in `.claude/settings.json` also runs this automatically when Claude Code opens.
  219. ## Staying current
  220. This repo updates regularly — new templates land in the prompt library, new workflows get added, bugs get fixed. To stay in sync with upstream:
  221. - **At every Claude Code session start**, the `check-context.sh` hook automatically runs `git fetch origin` (with a 10s timeout, never blocks). If your local clone is behind, the SessionStart banner will list the pending commits and tell you to run `git pull`. No surprise pulls — it just notifies.
  222. - **To pull updates manually:** `git pull origin main` from the repo root. If you've made local changes to tracked files, stash them first: `git stash && git pull && git stash pop`.
  223. - **If you've forked the repo on GitHub:** click the "Sync fork" button on your fork's page to bring it in line with this upstream, then `git pull` locally.
  224. - **Customizations:** your `.env`, `MASTER_CONTEXT.md`, `references/`, `outputs/`, and `logs/` are all gitignored — they survive every update. If you customize a core skill file (e.g. tune a SKILL.md for your brand), expect potential merge conflicts on `git pull` — keep custom versions under a non-tracked path (e.g. `local-skills/`) if you don't want them affected by upstream updates.
  225. ## Security
  226. - `.env` is gitignored — never committed
  227. - `MASTER_CONTEXT.md` is gitignored — contains your product IDs and workspace data
  228. - Never paste API keys in GitHub issues or public chats
  229. - Every Meta ad created via `meta-ad-builder` is created **PAUSED** — nothing goes live without you launching it manually in Ads Manager
  230. ## Vendor prompting guides
  231. | Model | Guide |
  232. |-------|--------|
  233. | Seedance 2.0 | Aligned to ByteDance's published Seedance prompting platform (the skill summarizes this in `skills/arcads-external-api/prompting/prompt-library/seedance-2.md`) |
  234. | Sora 2 | [OpenAI — Sora 2 prompting guide](https://developers.openai.com/cookbook/examples/sora/sora2_prompting_guide) |
  235. | Veo 3.1 | [Google Cloud — Veo 3.1](https://cloud.google.com/blog/products/ai-machine-learning/ultimate-prompting-guide-for-veo-3-1) |
  236. | Kling 3.0 | [Kling — user guide](https://kling.ai/quickstart/klingai-video-3-model-user-guide) |
  237. | Nano Banana | [Google Cloud — Nano Banana](https://cloud.google.com/blog/products/ai-machine-learning/ultimate-prompting-guide-for-nano-banana) |
  238. | ChatGPT Image 2 | OpenAI image-generation guidance (summarized in `shared/skills/chatgpt-image-ad/prompting/guide.md` with model-specific strengths and limits) |
  239. ## API docs
  240. [Arcads Swagger UI](https://external-api.arcads.ai/docs)
  241. ## Other AI assistants (Manus, Copilot, etc.)
  242. Point your assistant at [AGENTS.md](AGENTS.md) and `MASTER_CONTEXT.md` + the skill paths in `skills/` and `shared/skills/`. See [AGENTS.md](AGENTS.md) for details.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.