Novoads — AI Video & Image Ad Skill Pack

Every image and video in this README was generated by the skills in this repo, through the API, with
the asks shown beside them.

Make AI video ads and static image ads from Claude Code or Cursor, against your own
Novoads
account. The agent does the mechanical part — upload, price, generate, poll, download — and the
skills carry the part that decides whether the render is any good: the prompt.

Nine models are live on the API. Six make video (Seedance 2.0, Seedance 2.5, Seedance 2.0
Mini, Omni Flash, Veo 3.1, Sora 2) and three make stills (GPT Image 2, Nano
Banana Pro, Reve 2.1). On top of them this repo ships five Seedance prompt formulas, a
40-template static-ad library, Pixar and claymation pipelines, YouTube thumbnails, caption burn-in,
and a Meta publishing step.

Two rules run through the whole pack, and they are the reason it is safe to point an agent at a
billing API:

• No price is ever quoted from memory. Every credit number comes from a live
  POST /v1/estimates call in the session that is about to spend, shown to you and approved first.
  That call is free, and there is no rate table anywhere in this repo to fall back on.
• The spoken line is approved on its own, before the cost gate. Seedance renders the dialogue
  and the lip-sync in the same call, so the sentence in the prompt is the sentence in the finished
  ad. Approving a concept is not approving a sentence, and approving a sentence is not approving
  a spend.

Need an account? The entry offer is a $1 trial — not a free tier, and it can generate through
the API like any other live plan:
novoads.ai

Claude and Claude Code are products of Anthropic. This is an independent skill pack built by
Novoads for Claude Code; it is not affiliated with or endorsed by Anthropic.

Prerequisites

A Novoads API key, and a shell. The key is not optional and nothing in this
repo substitutes for it: every skill here runs one executable path, curl
against https://api.novoads.ai/v1 with NOVOADS_API_KEY from .env. Create
one at novoads.ai/dashboard/settings?tab=api.

Beyond the key, the core workflow (upload a photo, price it, generate a
video or an image, poll, download) is plain HTTP driven by your agent.
curl and jq are enough, and both are already on most machines. Your
first ad needs nothing from the optional table.

Captions, transcripts and music are generated server-side and need no local
tools: POST /v1/captions burns captions in (30 presets), POST /v1/transcripts
returns text, word timings and an SRT, and POST /v1/music generates the
soundtrack. Only laying that soundtrack under a finished cut is local, and that
is the music-mix step in the ffmpeg row above.

Optional, per workflow. Each skill tells you when it needs one; nothing
here blocks your first ad:

Linux: apt install curl jq ffmpeg python3. Windows: WSL2 or Git Bash;
the skills here are shell scripts and curl calls, so they need a POSIX shell.
No shell at all? See Environments that cannot run this repo at the bottom.

Get started (5 minutes)

Copy one prompt

In Claude Code or Cursor, open a new empty folder and paste:

https://github.com/novoads/claude-code-ads help me set this up

The agent clones this repo, runs ./scripts/setup.sh, and stops at the one step only you
can do: pasting your Novoads API key. Prefer the terminal? The same thing by hand:

1. Clone

git clone https://github.com/novoads/claude-code-ads.git
cd claude-code-ads

2. Run setup

./scripts/setup.sh

It creates .env (chmod 600), asks for your API key, and validates the key against
GET /v1/models before writing it to disk — a key that has never been probed is the most
expensive thing to debug later. It then copies MASTER_CONTEXT.template.md to your personal
MASTER_CONTEXT.md, syncs the skills, and runs the connectivity check.

Your key looks like novo_ followed by 64 hex characters. Create one at
novoads.ai/dashboard/settings?tab=api — it is
shown once, at creation, and cannot be retrieved afterwards.

Setup never hangs an agent: with no TTY (or with --non-interactive) it prompts for nothing,
prepares the workspace, prints the one step left for a human, and exits 0.

Re-check connectivity any time with ./scripts/check-novoads-env.sh. It tells you which failure
you have: a 401 is a bad or revoked key, a 403 with plan_required is a good key on an account
without API access. Different problems, different fixes.

3. Open in your editor

Claude Code: open the folder. A SessionStart hook syncs the skills and prints a banner —
which skills are installed, whether .env and MASTER_CONTEXT.md are set up, and the reminder
that prices come from /v1/estimates.

Cursor: open the folder. The same skills are exposed at .cursor/skills/.

4. Ask for something

Drop a product photo into references/products/ and say what you want. The agent picks the route,
reads the matching prompt formula, prices it, confirms the dialogue, and generates. The first time
you ask for an ad it asks what your product is and saves the answer to MASTER_CONTEXT.md, so no
later session asks again — setup itself never asks; the only thing setup needs from you is the key.

See what it makes

Everything below was made by this pack in one run on 2026-08-11/12, through the same skills a
customer installs. Each item shows the exact ask typed to the agent and the prompt the skill wrote
from it. The prompt panels are wrapped for width, and the long ones stop after their opening blocks
and name their full length.

Demonstration creative for a real product; not commissioned by or affiliated with the brand.

What	Watch	The ask, and the prompt the skill wrote
Pixar-style storyboard ad. A cast sheet, four beat stills, one clip per beat, voice-over into the gaps, a music bed and captions, assembled locally. Made with novoads-pixar-ad. Rendered 2026-08-12 UTC. Demo creative.		You type: Make a Pixar-style animated ad for Liquid Death — the villain is a sad crumpled plastic bottle. Play it as dark comedy, not eco-preaching. The prompt the skill wrote 3D animated film look, vertical 9:16, animating the character in the reference image. crushed unlabeled bottle, a clear plastic water bottle stamped flat and concertinaed along its length, lying on its side in hard ribbed folds, cap gone, two oversized half-lidded eyes and a small downturned mouth set into the rounded end, wrecked and never intact, no label and no printed text anywhere on it, lying in a shallow gutter puddle of black rainwater on wet asphalt at night under one amber street lamp. From zero to one and a half seconds the bottle sags deeper into the puddle and its eyes blink once, slowly. From one and a half to four seconds the eyes lift to camera and the bottle says, flat and defeated: "I've been in a cupholder since March." Small mouth shapes match the syllables and the eyebrows lift on the last word. Camera: locked extreme close-up macro, very subtle handheld micro-drift, no zoom, no pan. Stylized 3D animated feature film look, hard directional light with deep falloff, cold palette of wet asphalt grey, gunmetal blue, bone white and tarnished gold, painterly soft-focus background, desaturated colour grade with crushed blacks, every character caught mid-emotion. Audio: a small flat defeated male voice, close-mic and hollow, faint rain behind it. No other speech in this shot. The bottle is unchanged from the reference image: same crush creases, same eye placement. Exactly one bottle in frame. no live-action footage, no photorealistic humans, no uncanny faces, no dead eyes, no anime style, no 2D cel-shaded look, no named or copyrighted animated film characters, no morphing between frames, no on-screen text, no subtitles, no captions [beat 1 clip prompt, one of five. 1,677 characters, shown whole.]
Static ad to motion. One finished still, one call: the coin scratches the foil and the reveal happens while every printed word stays frozen. Made with novoads-image-to-motion. Rendered 2026-08-12 UTC. Demo creative. Rendered on seedance-2.0: on this run seedance-2.5 refused every start image it was sent, six calls, each charged and then refunded, while the same still went through on seedance-2.0.		You type: Animate this scratch-off ticket ad — a coin scratches the foil and the reveal happens. Every word stays frozen. The prompt the skill wrote Locked-off overhead flat-lay of a printed Liquid Death scratch-off ticket on grey painted concrete, a US quarter resting flat beside its lower-right corner. Photographic, not illustrated. A silent motion graphic with no spoken dialogue, no voice-over and no music. CAMERA: the camera is locked off and never moves: no pan, no tilt, no zoom, no push-in, no orbit, no drift. Lighting is constant: same soft key, same soft shadows, no light sweep, no flicker, no exposure change. The reference image is the exact composed state: every colour, shadow, layout position and printed glyph matches it precisely. Framing keeps the ticket and coin at their exact positions and relative sizes; nothing on the ticket is cropped or re-laid out. The ticket never moves, shifts, rotates, bends or lifts. TEXT: all printed text stays perfectly intact and pixel-identical to the reference at every frame, same faces, weights, sizes, navy and gold colours and positions, correctly spelled: "LIQUID DEATH PRESENTS", "No. 012777 / STILL ARTESIAN", "SCRATCH TO MURDER YOUR THIRST", "COLD", "MOUNTAIN", "ARTESIAN", "TALLBOY", "WIN", "MATCH ALL FIVE = ONE COLD CAN", "NO LOSING TICKETS. EVERY CAN IS A WINNER.", "liquiddeath.com", and on the small can photograph "Liquid Death" and "MOUNTAIN WATER". Text is never redrawn, re-lettered, re-spaced or re-coloured, and never fades, scales, slides or animates. The gold skull, the five navy icons and the can photograph are likewise pixel-identical and never redrawn. … 3,962 characters total
Claymation ad. A five-beat clay short: cast sheet, per-beat stills, per-beat clips, voice-over, music bed, a composited end card, captions burned locally. Made with novoads-claymation-ad. Rendered 2026-08-12 UTC. Demo creative.		You type: Make a claymation ad for Liquid Death — a bored office worker's grey day until the can shows up. No eco talk. The prompt the skill wrote Hand-sculpted stop-motion clay look. Plasticine characters with visible fingerprint impressions and sculpting-tool marks, matte clay surfaces with subtle micro-bumps, slightly asymmetric features, painted-on or sculpted eyebrows. Real knit-fabric clothing with visible wool weave and stitch lines, wooden and ceramic miniature-set props with hand-painted finishes. Practical miniature-set interior lighting, shallow macro depth of field, soft photographic bokeh. Subtle imperfection in every surface. Vertical 9:16 composition. Vertical 9:16, animating the character in the reference image. grey wool cardigan Frank, a man in his late 50s with thinning side-parted slate- grey plasticine hair in carved ribbon-strands and an oversized sculpted nose, stands in a miniature clay office cubicle set with grey felt partition walls in visible fibre texture, a hand-painted wooden desk in flat grey, a chunky beige clay desktop computer, a wire tray of hand-cut paper, a hand-thrown ceramic mug and a small clay desk lamp facing camera, the set behind him in saturated hand-painted colour. From zero to six seconds a small closed-mouth smile arrives on his face and he lifts the tall Liquid Death can to chest height with the label toward camera, one unhurried gesture. Camera: locked medium shot at chest height, very subtle handheld micro- drift, no pan, no zoom. The desk surface fills the lower third of the frame and stays clear of props throughout, fully rendered to the bottom edge of the frame. … beat 5 clip prompt, one of five. 3,807 characters total
UGC selfie ad. One 15-second Seedance call: four jump cuts, and the spoken script and its lip-sync rendered in the same call as the picture. Made with novoads-api. Rendered 2026-08-12 UTC. Demo creative.		You type: Make a 15-second UGC video ad for Liquid Death — a woman in her kitchen explains why she switched. Match Liquid Death's brand voice. The prompt the skill wrote 15 seconds UGC style honest review video, filmed on smartphone at arm's length, grey morning light through the kitchen window blinds, casual handheld selfie angle framed tight from the chest up. Grey hoodie woman: late 20s, dark hair in a messy claw clip, natural skin with visible pores across her nose and slight undereye shadows, wearing a plain unprinted grey hoodie with no writing or graphic on it. Her small apartment kitchen, close behind her and mostly out of focus — a plain white coffee mug, a wooden cutting board with a cut lime, a dish rack of clean plates. The setting stays the same throughout: the same kitchen counter, the window blinds as the only light source, the same plain grey hoodie and the same claw clip. Only the framing changes between cuts, and every cut stays close — her head and shoulders fill the frame and the can is never further from the lens than her own face. The video opens on a tight chest-up selfie framing — grey hoodie woman looking flatly into the lens, holding the @Image1 tall white and navy Liquid Death mountain water can up beside her chin, close to the camera: "I switched to water. My family staged an intervention." Quick jump cut — grey hoodie woman closer still, the @Image1 can filling half the frame height with its front face square to the lens: "It's called Liquid Death. It's water. Terrifying." Quick jump cut to a tight framing — grey hoodie woman takes one flat unimpressed sip from the @Image1 can, the can still large in frame: "Tastes like water. That's the whole review." Quick jump cut — grey hoodie woman holds the @Image1 can up beside her face, both filling the frame: "Murder your thirst. Or don't. Whatever." … 3,404 characters total
Clone of an existing ad. One of our own skincare ads read locally with ffmpeg and Whisper, then rebuilt for a different product. The clip plays source left, clone right. Made with clone-video-ad. Rendered 2026-08-12 UTC. Demo creative.		You type: Clone this video ad for Liquid Death: references/sources/our-own-ugc-ad-source.mp4 The prompt the skill wrote Handheld selfie take on a phone in one hand, one unbroken take, no cuts, framing drifting as her arm moves. A woman in her late twenties, damp dark hair half pulled back, bare face, visible pores, freckles across her nose, white waffle bathrobe, bright bathroom, mirror behind her, blown-out window at camera-left. [00:00] Chest-up, she looks into the lens, flat little smile, and says: "Okay, my thirst has genuinely never been this dead." [00:04] She raises the @Image1 tallboy can into frame beside her jaw and says: "Just finished my morning water and I'm kind of obsessed." [00:07] She slowly turns the can toward the lens, front label to camera, and says: "Nothing sweet, nothing added, just water and aluminum." [00:10] She takes one slow sip and lowers the can to her collarbone, saying: "Honestly, it's the only thing in my routine that isn't lying to me." [00:13] Back to chest-up, the can still at her collarbone, deadpan half- smile, she says: "All right, ready to murder some thirst." The can from @Image1 stays visually unchanged in every shot, its label sharp and identical to the reference image, its text unchanged and fully legible, with no extra words, numbers or lettering anywhere on the can. Robe and damp hair unchanged throughout. The tone is dry, warm and flat; she talks quickly, running sentences together like a voice note, every word landing clearly. Documentary handheld, photorealistic, natural phone quality, no colour grading, one side of her face in shadow, sound direct from the phone mic, room echo. Vertical 9:16. [1,555 characters, shown whole.]

Five templates from the
40-template library, plus a
thumbnail, on one sheet: scratch ticket, desk-kit flatlay, 9:16 story hero, sticky-note flatlay,
before/after split, and a reaction thumbnail from
generate-youtube-thumbnail.

The pair demonstrates the edit endpoint: the same image sent back with sourceAssetId, one new
headline, and nothing else moved.

The pack prices before it spends: POST /v1/estimates first, the number shown, the render only
after that number is approved. This is what one of those moments looks like, captured on 2026-08-11
before the UGC row's 720p render. A dated capture, not a rate table.

{
  "capturedAt": "2026-08-11",
  "endpoint": "POST https://api.novoads.ai/v1/estimates",
  "apiSpecVersion": "2.19.0",
  "httpStatus": 200,
  "request": {
    "kind": "video",
    "model": "seedance-2.0",
    "durationSeconds": 15,
    "resolution": "720p",
    "language": "en",
    "prompt": "15 seconds UGC style honest review video, filmed on smartphone, grey morning light through the kitchen window blinds, casual handheld selfie angle. … (2,646 characters)"
  },
  "response": {
    "credits": 7,
    "balance": 243.2,
    "sufficient": true
  },
  "responseHeaders": {
    "x-request-id": "f8f1379240ff64680b8263fa8d5559b3"
  }
}

credits and balance are display credits, the same unit POST /v1/videos reports back as
creditsCharged. The prompt is abridged here for width; the call carried it in full.

What you can make

🎬 Seedance 2.0 video — the flagship

Any integer duration from 4 to 15 seconds, six aspect ratios, native audio and lip-sync
rendered in the same call. Start from one photo (startImageAssetId animates it as the first
frame) or composite up to nine reference images (referenceAssetIds, addressed in the prompt
as @Image1…@ImageN). Those are two separate modes — a request carrying both is rejected.

Five prompt formulas ship, each with a worked example that has been priced live:

"Make a 12-second Seedance UGC video — woman in a kitchen holding the product, says she stopped
buying the drugstore version"

The platform guide — grids, the two reference modes, prompt length by shot count, the mini-draft
tier — is seedance-2.md.

Draft on Mini first. seedance-2.0-mini takes the same grid and the same formulas at half the
price and comes back in 2–3 minutes. Upload once, draft on Mini, re-price naming the final tier,
render once. The agent asks which tier you want before the first Seedance call of a workflow.

🎬 Seedance 2.5 — the long one

Any integer duration from 4 to 30 seconds, which makes it the only model here that renders past
15 in a single call. Everything else is 2.0's shape: the same six aspect ratios, the same nine
referenceAssetIds, the same audioEnabled toggle, the same 4,000-character prompt ceiling — so
the five Seedance formulas apply unchanged.

Two differences worth knowing before you pick it. It renders 480p and 720p only — no
1080p, no 4k, on either provider — so a resolution carried over from a 2.0 workflow is a
rejected request rather than a downgrade. And it is the dearest model on the API at the same
length; thirty seconds of it is the most expensive single call this API takes. Price it, like
everything else, at POST /v1/estimates.

"Make a 30-second Seedance 2.5 ad — the founder walks through the workshop and explains why they
built it"

A longer clip wants more beats, not slower delivery. If the script only fills 15 seconds, render
15 seconds on seedance-2.0, which is cheaper at that length.

🎬 Omni Flash — fast vertical clips

"Give me a fast 8-second vertical clip of the product on a kitchen counter, no dialogue"

omni-flash carries a 20,000-character prompt ceiling, the roomiest of any video model here
against 4,000 everywhere else, which is what makes long structured briefs possible. Its grids are narrower than Seedance's — durations 4/6/8/10,
aspect ratios 9:16 or 16:9 only — and it takes no reference images. Prompt craft in
shared/skills/gemini-omni-flash/prompting/guide.md,
which is Google DeepMind's own guide with the parts this API cannot reach marked as such.

🖼️ Stills — people, products, characters

POST /v1/images is synchronous: the finished images come back in the response body, so there
is nothing to poll. Up to 4 images per call; reference images are capped per model — 4 on
gpt-image-2, 14 on nano-banana-pro, 8 on reve-2.1.

• gpt-image-2 — heavy typography and mimicked UI.
• nano-banana-pro — photoreal people and products in a scene; holds a character's identity
  tightest across a reference batch.
• reve-2.1 — a different look on the same still, or a second opinion.

"Create a new AI influencer — 22-year-old with freckles, golden-hour kitchen lighting"
· "UGC selfie of Sofia holding the product in her bedroom"
· "Recreate this influencer's look from this photo"

Character sheets, influencer recreation, UGC product selfies and product-showcase stills live in
skills/novoads-api/prompting/prompt-library/. The
standard flow is still-first: generate the still, get it approved, then animate the approved frame
on Seedance. Stills are cheap to iterate; video is not.

Every returned image gets a QA look — hands, fingers, faces, merged objects, garbled text. A
defect is regenerated with a corrected prompt, capped at 2 retries, and the extra credits are
reported at the end.

📸 Static Meta image ads (40-template library)

"Make me an Apple Notes-style ad for my product" · "Generate a Forbes editorial ad" ·
"Clone this comparison-table ad as a template"

A three-skill family over a shared library of 40 validated prompt templates — Apple Notes
lists, editorial hero, fake Google search, comparison tables, sticky-note flatlays, Slack threads,
ChatGPT-conversation ads, iMessage screenshots, magazine covers, billboards, museum exhibits,
weather UI, scratch-off tickets, founder letters, dating-app cards, and more.

• chatgpt-image-ad — typography / UI mimicry, on gpt-image-2
• nano-banana-image-ad — photoreal / lifestyle / multi-reference, on nano-banana-pro
• clone-image-ad — reverse-engineer an existing ad into a new library entry; asks which backend to validate against at Phase 1 (gpt-image-2, nano-banana-pro or reve-2.1) and can cross-validate against another at Phase 8

Read shared/skills/image-ad-prompting/OVERVIEW.md
first — it holds the decision tree, the per-backend aspect-ratio matrix, and the standard
generate / clone workflows. Output is image files; publishing is the separate meta-ad-builder
skill.

🎞️ Multi-step animated pipelines

🔄 Reverse-engineer existing creative

"Reverse-engineer this video into a reusable Seedance template" ·
"Clone this video ad for our product"

• analyze-video — pulls frames with
  ffmpeg and the transcript with Whisper locally, reads out the beat structure, and writes a new
  parameterized formula into the prompt library. Nothing is charged until an optional test render.
• clone-video-ad — the same local analysis, but the
  output is a rendered clip, so both gates apply. There is no video-to-video on this API: a source
  longer than the chosen model's ceiling becomes a series of clips held together by passing the same
  reference images to each one — though seedance-2.5 reaches 30s in one call, so some sources that
  used to need two clips now need one.

📤 Publish as a paused Meta ad

"Publish this approved creative as a paused Meta ad"

meta-ad-builder takes a finished file and uploads it
through the Meta Marketing API. Every ad is created PAUSED — you launch it yourself in Ads
Manager. It can also pull your top-spending ads and competitor ads to inform copy. Auth via the
META_* rows in .env.

What it costs

There is no price list in this repo, on purpose. A stored rate goes stale silently, and a quote
that disagrees with the invoice is worse than no quote at all.

Instead, before anything is generated, the agent calls POST /v1/estimates and shows you what came
back: the credits for that exact configuration, your balance, and — when the batch would exceed the
balance — how short you are and where to top up. You approve the number, then it generates.

The estimate is free, and what it does not do is worth stating:

• The estimate lints a video prompt, advisorily. It returns a warnings array of craft notes
  ({rule, message}) — verified live 2026-08-04. POST /v1/videos and POST /v1/images do not
  run these rules and return no such field. Nothing in it can refuse or reprice a call, so a
  weak prompt is still priced, charged and rendered exactly like a strong one, and the warnings
  are substring matches that do false-positive. Read them, judge each against your prompt,
  and never paste in a suggested fix that does not fit. The prompt libraries in this repo remain
  the real quality gate — which is why the skills treat them as mandatory reading rather than
  reference material.
• It refuses malformed requests for free, including a prompt over the named model's character
  ceiling.
• Its quote cannot disagree with the invoice, with one caveat worth knowing: the estimate skips
  moderation, which the paid call runs, so a prompt priced clean can still be refused as
  content_policy. Nothing is charged when it is.

GET /v1/models carries each model's live grid and its price for a representative output if you
want to compare models yourself.

Supported models

All nine are live on api.novoads.ai/v1. Grids below come from GET /v1/models — that endpoint
is the current answer, this table is a map.

Image calls take numImages 1–4. Two of their limits are per model and neither is one
number across the set. The referenceAssetIds cap is 14 on nano-banana-pro, 8 on
reve-2.1, 4 on gpt-image-2. The prompt ceiling is 50,000, 4,000 and 32,000 in that same
order, split apart by deployed spec 2.16.0 from what used to be 4,000 everywhere. So a prompt
written for one image model may be too long for another, and switching to reve-2.1 for a
second opinion is the case that hits it. Images have no start-frame concept. Videos are
asynchronous (202 + jobId, poll to a terminal status); images come back in the response body.

audioEnabled is a Seedance-only boolean (default true) — all three variants take it; send
false for a clip meant to run silent. The other three video models are strict and 400 on it, as
does POST /v1/estimates for every model — it does not move the price.

resolution is on seedance-2.0 and seedance-2.5 only, and it is the one output-shape field
that moves the price: each tier is its own credit schedule, not a surcharge on the one below it.
Since 2026-08-07 480p costs roughly half of 720p rather than the same, which makes it a real
draft tier on both. seedance-2.5 stops at 720p.

Kling 3 is not on this API and is not queued for it. Its prompt library sits in
skills/novoads-api/prompting/prompt-library/ as craft only; the agent will say so plainly rather
than routing you somewhere else. There is likewise no b-roll or scene endpoint — a silent
omni-flash or Seedance clip is the b-roll path.

What's in the box

Reference images

Drop images into references/ and the agent finds them before asking you for anything:

• references/influencers/ — faces and bodies to recreate, plus saved character sheets
• references/products/ — product photos for showcase and UGC workflows
• references/aesthetics/ — mood boards, lighting and style references

The folder is gitignored, so the files stay on your machine. Uploads are minted through
POST /v1/uploads and the resulting assetId is durable and reusable forever — upload the
product shot once and reuse it across every iteration and every model, which is what makes drafting
on Mini cheap. (The presigned upload URL expires in 900 seconds; the assetId does not.)

Project memory

MASTER_CONTEXT.md is your workspace's living memory — created by setup, gitignored, read at the
start of every session and written back to. It holds:

• Default product — auto-populated on first use, so you are never asked "which product?" twice
• Brand voice — tone, audience, words to use and avoid
• API learnings — empty on purpose, filled by your own sessions with what the skill cannot know
• Meta ad defaults — account, page, pixel IDs for meta-ad-builder
• Changelog — dated entries as decisions accumulate

It holds no prices, deliberately. See what it costs.

Editing skills

Canonical sources live under skills/ and shared/skills/. After editing any file there:

./scripts/sync-skill.sh

That copies changes into .claude/skills/ and .cursor/skills/, which are generated and
gitignored. The SessionStart hook runs it automatically when Claude Code opens the folder.

Staying current

• At session start, the banner hook runs git fetch origin (10s timeout, never blocks). If your
  clone is behind, it lists the pending commits and tells you to pull. No surprise pulls.
• To update: git pull origin main. With local edits to tracked files, git stash && git pull && git stash pop.
• Your data survives updates: .env, MASTER_CONTEXT.md, references/ and outputs/ are all
  gitignored. logs/ is not — latency and failure history across sessions is worth keeping, which
  is exactly why the log never records keys, presigned URLs, or prompt text.
• Customizing a shipped skill file invites merge conflicts on pull. Keep heavily customized
  versions under a non-tracked path if you would rather not deal with them.

Security

• .env is gitignored and written chmod 600. Never commit it.
• MASTER_CONTEXT.md is gitignored — it holds your product IDs and workspace data.
• The agent never prints your key back to you, and never writes it into MASTER_CONTEXT.md. If you
  paste a key into a chat, rotate it — transcripts get shared.
• Keys are shown once at creation and are scoped to the organization that created them. Another
  organization's object returns 404, not 403: "not found" and "not yours" are deliberately
  indistinguishable.
• Every Meta ad created by meta-ad-builder is PAUSED. Nothing goes live without you.

API docs

• docs.novoads.ai — the full API reference (Scalar).
• GET /v1/openapi.json — the machine-readable spec.
  Public, no key needed, and the authority whenever a file in this repo disagrees with it.

Environments that cannot run this repo

Everything above needs a POSIX shell, curl, and a key in .env. If you have none of that
(claude.ai on the web, or Windows with no WSL and no Git Bash), Novoads also runs a hosted
connector at novoads.ai/mcp. It is a separate surface, not a
way to run this pack: it carries none of the skills, prompt libraries or formulas in this repo,
and it quotes costs in different units. The skills in this repo never use it. They stop and
ask for the API key instead, even in a session where the connector is already connected.

Vendor prompting guides

Other AI assistants (Manus, Copilot, etc.)

Point your assistant at AGENTS.md — it carries the auth shape, the async-video /
sync-image distinction, the cost policy, and the skill map.