@koda-sl/baker-cli 0.424.0 → 0.425.0-dev.107cc9af3

package/README.md

@@ -1803,6 +1803,20 @@ baker gsc sitemaps --site-url "https://example.com/"
 
 ---
 
+### Making ads (`baker creatives brief` / `ideas` / `make`)
+
+Ads start from what customers said. Three commands carry the process; the agent writes the ideas and the copy.
+
+```bash
+baker creatives brief [--adset <id>] [--account-id act_<id>]   # one read: viewer, account bottleneck, what works, taste, customer quotes, ideas on file
+baker creatives ideas [--full]                                 # check and rank src/creatives/_ideas/*.md; build_next
+baker creatives make <idea> [--slug <s>] [--adset <id>] [--reference <adId>] [--spec <beats.json>]
+```
+
+- `brief` returns a verdict, not raw numbers: `account.bottleneck` is `hook`, `after_click`, `none` or `too_little_data`. A part that cannot be read says `unavailable`.
+- `ideas` is free and local. An idea's `evidence` resolves to `src/content/insights/` entries, or it is an `explore` bet. Numbers must be in the Knowledge or our own customers' quotes (a Reddit, forum or web quote never sources a claim), no two ideas may share an angle and format, and a review or testimonial format needs a quote from our own customer.
+- `make` builds one idea. A static searches the idea's container for ads made for our viewer, checks the closest three with `winning-ads fit` and scaffolds from the first that fits. A video idea takes its beats spec. Either way it records the idea on `_definition.md`, sets `prompt.json`'s `message` and syncs the definition. Drafted copy lines claiming a figure the company never wrote come back first in `hints`.
+
 ### Competitive Intelligence (`baker research`)
 
 Market and competitor research powered by DataForSEO and AI-powered Google Search grounding. Shows who's competing for keywords, search intent classification, competitor keyword strategies, keyword gaps, landing page performance, and open-ended AI research.
@@ -1934,6 +1948,16 @@ Every search or thread item carries verbatim `text`, `author`, `score`, `created
 
 ---
 
+### `baker research customers`
+
+Finds what real customers say and writes each quote as an insight in `src/content/insights/` (type, quote, source, date, `status: observed`). Sources: the company's reviews library, Reddit, and with `--web` pages from web research. A quote is kept only when it appears word for word in the text that was fetched; the model classifies quotes and never writes them.
+
+```bash
+baker research customers --focus "solar panels"     # reviews + Reddit
+baker research customers --query "payroll mistakes" --web --limit 15
+baker research customers --dry-run                   # print, write nothing
+```
+
 ### `baker research advertisers "keyword"`
 
 Returns a ranked list of **domains** competing for one keyword in Google, with average SERP position, relevance rating, estimated traffic value and visibility — domain economics, **NOT ad copy**. Despite the name it returns no headlines, descriptions or creative of any kind; for a competitor's actual ad copy use `baker winning-ads content <adId>` (Meta/LinkedIn only — Google SERP ad copy is not available anywhere in Baker).
@@ -3267,6 +3291,18 @@ next one that shares her name. A `--handle` you pass yourself still conflicts: t
 decision, and moving it quietly would point `@marta` at somebody else. The returned `handle`
 is the one the avatar actually got.
 
+#### `baker avatars research "<who>"`
+
+Find a real, named person — a testimonial author, the company's founder, a team member — from a description, and gather their public photos into the image library. `--create` builds their avatar from those photos right away (likeness `licensed`, no face picker), or returns the avatar they already have.
+
+```bash
+baker avatars research "Lautaro Suarez, CTO at Vibiz"
+baker avatars research "the CEO of Vibiz" --create
+baker avatars research "Lautaro Suarez" --company Vibiz --role CTO --create --handle lautaro
+```
+
+The answer's `status` is `found` (their `person` record, `photos`, a `subjectDescription` and, with `--create`, the `avatar`), `possible` (`candidates` that matched only in part — pick one with `--linkedin <url>`) or `not_found` (the `searched` queries; nothing was taken). A profile counts as the person only when both the name and the company match it. It checks the company's avatars first: a person who already has one (the same full name, and the same company when both name one; for a role alone, the researched avatar holding it) comes back `found` with that `avatar` (`reused: true`) — nothing searched, billed or copied again; an invented face (likeness `synthetic`) or one whose build failed never counts as them. `--linkedin` skips that check. When a lookup service does not answer (a timeout, a rate limit), the command fails with `UNAVAILABLE` rather than answering `not_found`, and nothing of that run is kept, so the next try searches again. `--name`, `--role` and `--company` override the description; `--full` adds their about text, location and every search run.
+
 #### `baker avatars update <handle>`
 
 Edit an avatar. Profile flags are merged over the stored profile, so `--persona` alone keeps the subject description intact; an empty value clears the field it names (except `--subject`, which can only be replaced). Changing `--subject` requires `--regenerate-sheet` in the same call — otherwise the sheet still shows the person described before.
@@ -3415,6 +3451,27 @@ baker winning-ads formats --awareness solution_aware --output md
 Each key returns `count`, `mean_winner_score`, and `winner_share`. Ranking is winner-weighted volume, so a
 proven grand-slam pattern floats to the top and a measured dud is floored.
 
+### `baker winning-ads fit <adId>`
+
+Before building on a reference: why the ad wins, read against **this** company. One billed vision call, cached per ad
+under `canvas/.cache/fit/<platform>-<adId>.json` (asking again is free).
+
+```bash
+baker winning-ads fit a_12345 --platform meta
+```
+
+Returns `why_it_wins` in five layers (`disguise`, `voice`, `trigger`, `attention`, `logic`), the `load_bearing`
+elements that carry them (each with what it `needs`, whether the company has it: `have` | `adapt` | `missing`, and
+`ours`, what replaces it), and a `verdict`: `use`, `adapt` or `reject`. The company's Knowledge, reviews and market
+ground the answer. `baker canvas scaffold-static-ad --source-ad-id <adId> --source-platform <platform>` reads the cached
+answer, writes it into `prompt.json` as `concept`, and refuses a `reject` with `concept_does_not_fit` before anything is
+billed; the hint names the platform the fit was asked on, so a TikTok or LinkedIn reference is read from its own library.
+
+| Flag | Default | Description |
+|---|---|---|
+| `--platform <meta\|linkedin\|tiktok>` | `meta` | Which platform the id is on |
+| `--full` | off | The whole answer, with the reasoning per element |
+
 ### `baker winning-ads advertisers <brand>`
 
 List corpus brands by name or domain → `GET /api/ad-library/advertisers`. Use it to find **your own** advertiser (to `--exclude-advertiser`) or a **competitor** (to `--advertiser-id` / `winners`). Lean default fields: `advertiser_id`, `label`, `platform_count`, `family_count`, `active_ad_count`, `total_ad_count`; `--full` adds `scraped_name`, `image_url`, `platforms`, `total_reach`, `last_synced_at`. Supports `--platform meta|linkedin`, `--limit`, `--offset`.
@@ -4572,7 +4629,7 @@ Local; no credits (one box-reader call per variant, about $0.005, which rides on
 
 ##### `static_safe_fit`
 
-Local; no credits (one box-reader call, about $0.005, which rides on the run record's review spend). Brings a finished 9:16's words and logo inside the band Reels or Stories leave clear without rendering again: reads where every line and the logo sit and, when any reaches past the band, shrinks the whole picture a little (to 85% at most) and shifts it, filling the uncovered margin by mirroring the picture beside it, blurred under a feathered seam. The image model cannot move an element by a few points on request, so this replaces re-rendering for small misses. **Inputs:** `image`. **Params:** `format`, `placements` (the ad set's, as the sizes were planned). **Outputs:** `image` (same pixel size), `fit` (`{ scale, x, y }`, or null when passed on unchanged), `reason` (`not_read`, `inside_band`, or `too_far` when clearing the band needs more than the 85% shrink: the layout itself has to change) and `usage`. Scaffolded by `scaffold-static-ad` after every 9:16 an interface covers (`safe_9x16`, tagged `static_safe_fit`).
+Local; no credits (one box-reader call, about $0.005, which rides on the run record's review spend). Brings a finished 9:16's words and logo inside the band Reels or Stories leave clear without rendering again: reads where every line and the logo sit and, when any reaches past the band, shrinks the whole picture a little (to 85% at most) and shifts it, carrying its edge into the uncovered margin, only when that edge is plain ground; where the review blocks a line under Reels' or TikTok's buttons down the right (an ad set that runs there and not in Stories), the fit clears them too (a busy edge would read as a blurred duplicate, so the picture is passed on as rendered). The image model cannot move an element by a few points on request, so this replaces re-rendering for small misses. **Inputs:** `image`. **Params:** `format`, `placements` (the ad set's, as the sizes were planned). **Outputs:** `image` (same pixel size), `fit` (`{ scale, x, y }`, or null when passed on unchanged), `reason` (`not_read`, `inside_band`, `too_far` when clearing the band needs more than the 85% shrink, or `busy_edge` when the edge it would carry is not plain ground) and `usage`. Scaffolded by `scaffold-static-ad` after every 9:16 Reels or Stories covers (`safe_<size>`, tagged `static_safe_fit`). It never sets the picture over a blurred or mirrored copy of itself: a layout `too_far` or `busy_edge` is passed on as it is, the run review names what Reels or Stories covers, and `baker canvas run` lays that size out again by itself.
 
 ##### `font_specimen`
 
@@ -5774,7 +5831,7 @@ cart, a trial, a donation, a registration); who it is for is reported but never
 `_definition.md`, else the one its `targetAdSet` reaches (as `baker ads meta adsets --audience` reads it), else
 the persona and `awareness` from the Knowledge, else a cold stranger. Not understood → exit 2 with
 `viewer_check_failed` and a `fix` naming what the viewer missed. Answers are cached per copy, scene and viewer.
-`baker canvas run` makes the same check on a static ad before it spends (skipped for `--fix` and resumed runs;
+`baker canvas run` makes the same check on a static ad before it spends (skipped for `--fix`, an autofix round, `--use-variant`, a re-layout and resumed runs, never for an edit of the words;
 `--skip-viewer-check` renders without it; a model outage lets the render through with a warning). `--full` adds
 the viewer's whole answer and how the viewer was chosen.
 
@@ -5802,9 +5859,10 @@ A completed render the review calls ready (`verdict` `ready` or `ready_with_note
 | `--concurrency <n>` / `--parallel <n>` | `8` (or `BAKER_CANVAS_CONCURRENCY`) | Max nodes executing at once within a layer. |
 | `--fresh` | off | Start a new run id instead of resuming an interrupted run of this canvas. |
 | `--regenerate <ids\|roles>` | none | Force these nodes, and everything downstream, fresh this run only. A role names every node carrying it: on a static ad `static_variants` (or its id `gen`) renders the variants again. An unknown name exits 2 (`unknown_regenerate_node`). |
-| `--fix "<what to change>"` | none | Static ad only: correct a finished version with one render that changes only what you say; every size is made again from it (see below). One render of the round. |
+| `--fix "<what to change>"` | none | Static ad only: correct a finished version with one render that changes only what you say; every size is made again from it (see below). The same fix after it left its defect in place is refused (`no_progress`). |
 | `--use-variant <sha256>` | none | Static ad only: make every size again from another of a version's variants, the one the person chose; the only requests re-lay it out for each other size (see below). |
 | `--from <run_id>` | the last render | With `--fix` or `--use-variant`: the version to start from. |
+| `--autofix` / `--no-autofix` | on | Static ad only: when the review finds a blocking defect, make up to 2 more renders (`AUTO_FIX_ROUNDS`) that correct it before reporting (see below). `--no-autofix` reports after the first render. |
 | `--skip-viewer-check` | off | Static ad only: render without the viewer check (`baker canvas viewer-check`), when the person heard what the viewer would miss and wants the copy as it is. |
 | `--remote-cache <on\|off>` | `on` (or `BAKER_CANVAS_REMOTE_CACHE`) | Operator only. Company-scoped remote cache + durable asset persistence. |
 | `--no-record` | records | Skip posting the durable run-history record (and its live progress). |
@@ -5819,7 +5877,9 @@ A completed render the review calls ready (`verdict` `ready` or `ready_with_note
 
 **Resuming an interrupted run.** A long `baker canvas run` (multi-clip video) that is killed mid-render — session end, sandbox pause — leaves a marker under the outputs dir. The next `baker canvas run` of the same canvas automatically **resumes** that run: it reuses the run id so still-running billed jobs re-attach instead of being abandoned and re-billed, and completed nodes come from the cache. A live render keeps its marker's heartbeat fresh, so a second `run` of the same canvas is refused while it is alive (checked with `kill(pid, 0)` plus the heartbeat, on Linux and macOS alike). Resume also works from a **different workspace or a fresh sandbox**: when no local marker survives, the run history is consulted for a resumable run of the same source (canvas, `prompt.json`, scenes, compositions and formats), and it is adopted when the backend says it may be: interrupted, stopped at a checkpoint, or a `running` row whose keepalive is older than the backend's staleness threshold. Ctrl-C / SIGTERM aborts gracefully — no new nodes dispatch, a resumable snapshot is flushed, and the marker survives. A clean completion (or a handled failure) clears the marker, so a normal re-run starts a fresh generation. Force a new run with `--fresh`, or pin a specific run with `--run-id <id>` (also the escape hatch to adopt a run that is reported as concurrently live). Independent same-layer nodes (e.g. video clips) fan out in parallel up to `--parallel`/`--concurrency` (default 8; env `BAKER_CANVAS_CONCURRENCY`).
 
-**Static ads: correct a version, or use another of its variants.** Correct a finished version with `--fix "<what to change>"`: one render takes that version (the last one, or `--from <run_id>`) as its first reference and changes only what you said, keeping the picture, the layout and every other word; it replaces the variants and the judge, every size is made again from the corrected picture, and it counts as one render of the round. Use another variant with `--use-variant <sha256>` (`--from <run_id>`, default the last render): when the person chose another of a version's variants over the judge's pick, every size is made again from that picture, and the only requests re-lay it out for each other size; `--regenerate adapt_9x16` lays out one size again from the same chosen picture when only that size came out wrong (after a `--fix` or `--use-variant`, from that corrected or chosen picture, so the size never brings back the uncorrected one; `--from <run_id>` names another version); `baker creatives feedback --slug <slug>` prints the choice (`variantChoice`) and the exact command.
+**Static ads: the run corrects a blocking defect itself.** When the review of a static ad finds a blocking defect, `baker canvas run` makes up to 2 more renders (`AUTO_FIX_ROUNDS` in `static-autofix.ts`) before it reports, correcting each defect at the layer that causes it: words or a logo where an interface covers them (Reels, Stories) or the frame cuts them are **placed and extended** (`withExtendedSizes`): a `static_place` node reads where everything sits and shrinks the ad (to 60% at most) into the size's safe area, and a `static_extend_merge` node has the ad's own model paint only the margins that leaves and lays the ad's own pixels back over the painting, so the model is never asked to move anything (a size already inside its area is passed on with nothing painted or billed, and one whose words sit too far out for a shrink to clear is laid out again instead); a defect the spec asked for (a second subject, stars with no named source, a concept that does not read) and any defect of a native screen corrects the spec first (`static-spec-repair.ts`: a model rewrites `prompt.json` for those findings alone, every line of the copy kept word for word), then draws the ad again from it (a native screen redraws free); a picture wrong as a whole or in three places or more is made again from its brief, told what the last one got wrong, but only in a round with another after it unless the brief caused the defect; anything else on the picture is a `--fix` of the chosen picture for its own findings (`correctionOf`: the words to erase, the line to re-letter or add), naming what the last edit was already asked for, while each re-laid size follows as an edit of its own last layout (`withSizesFollowingFix`), keeping what an earlier round fixed in it; anything else on a re-laid size alone edits just that size from its own last layout (`withRelayEdit`). A size a round did not touch (its file the same byte for byte) keeps the round before's verdict (`carryUnchanged`), so the review's second look at the same pixels never counts as a new defect. Each round is one render (a fix, or one request per re-laid size) and prints `[autofix] round N of 2: …` on stdout; the final JSON's hints start with "Corrected automatically before this result, …" and its review is of the corrected version. Blocking findings left after the rounds keep the hints below. A round that would only repeat the correction that just left its defect in place (a chosen option, `--use-variant`, is only ever corrected) is not run: that version is reported as it is. `--no-autofix` reports after the first render.
+
+**Static ads: correct a version, or use another of its variants.** Correct a finished version with `--fix "<what to change>"`: one render takes that version (the last one, or `--from <run_id>`) as its first reference and changes only what you said, keeping the picture, the layout and every other word; it replaces the variants and the judge, and every size is made again from the corrected picture. **A `prompt.json` changed only in its words, their positions or its colours** since the version shown is made the same way, as an edit of that version (`[edit]` in the output, the instruction read off the change); three new options are asked for only with `--regenerate static_variants`. A `--fix` whose defect is still blocking afterwards is recorded, and asking for it again with the same words exits 2 with `no_progress`. When every blocking finding is on re-laid-out sizes, the first hint says to lay out those sizes again (`--regenerate adapt_<size>`) before any `--fix`. Every static run's hints say what it cost in credits and what for. There is no render limit per creative. Use another variant with `--use-variant <sha256>` (`--from <run_id>`, default the last render): when the person chose another of a version's variants over the judge's pick, every size is made again from that picture, and the only requests re-lay it out for each other size; `--regenerate adapt_9x16` lays out one size again from the same chosen picture when only that size came out wrong (after a `--fix` or `--use-variant`, from that corrected or chosen picture, so the size never brings back the uncorrected one; `--from <run_id>` names another version); `baker creatives feedback --slug <slug>` prints the choice (`variantChoice`) and the exact command.
 
 #### `baker canvas scaffold-video <video> [flags]`
 
@@ -5930,24 +5990,24 @@ The two scaffold passes are billed (the full `video_deconstruct` is the heavy on
 
 Turn a source/inspiration image into a **runnable, self-validated static-ad canvas** — the static counterpart of `scaffold-video`. It runs **billed Gemini passes** up front:
 
-1. **`image_describe`** (`~google/gemini-pro-latest`) — reverse-engineers the image (layout by what fills each region, copy verbatim, typefaces, capture in words, the `mechanism` and the `audience` it is built for) and writes it next to the canvas as **`prompt.json`**, an **ad spec** (`"schema": "baker-static-ad/2"`) you rewrite into the ad you want: `viewer` (who sees it and what they already know), `message` (the one thing it tells them, never printed), `copy` (each line with `lettering`: `in_scene`, written on something in the picture (`on`, `look`), or `designed`, set in the brand's face), `scene` (`description`, `look`: photo | designed | illustration, `capture` in words, `mood`, `layout`), `brand`, `colors` (each tied to an object, from the brand palette), `references` (one picture each, labelled by role: `PERSON_1`, `ANIMAL_1`, `PRODUCT_1`, `BADGE_1`, `PROP_1`), `logo` (and `logo_position`), `third_party_brands` (other brands the offer is about, a car brand on a car loan: `name`, `what`, and `logo` naming the `BADGE_n` reference of its logo; their logos may appear), an optional `direction`, and the `mechanism` to rebuild for our brand. **You never write a model prompt**: every render prompt is compiled from this file for the model it runs on (short labelled sections for gpt-image, sentences with reasons for Gemini; each line quoted once; pictures numbered with one role each; colours from the brand; realism as capture cues; the safe area from its one owner; constraints last), on every `validate` and `run`, and shown on its node. A `prompt.json` written before the ad spec (the describe blueprint) is read through an adapter and keeps working. The source's own read (`source`, and the library's `reference`) stays as bookkeeping and is never sent to a model.
+1. **`image_describe`** (`~google/gemini-pro-latest`) — reverse-engineers the image (layout by what fills each region, copy verbatim, typefaces, capture in words, the `mechanism` and the `audience` it is built for) and writes it next to the canvas as **`prompt.json`**, an **ad spec** (`"schema": "baker-static-ad/3"`; `/2` still reads) you rewrite into the ad you want: `viewer` (who sees it and what they already know), `message` (the one thing it tells them, never printed), `copy` (each line with `lettering`: `in_scene`, written on something in the picture (`on`, `look`), or `designed`, set in the brand's face), `scene` (`description`, `look`: photo | designed | illustration, `capture` in words, `mood`, `layout`), `brand`, `colors` (each tied to an object, from the brand palette), `references` (one picture each, labelled by role: `PERSON_1`, `ANIMAL_1`, `PRODUCT_1`, `BADGE_1`, `PROP_1`), `logo` (and `logo_position`), `third_party_brands` (other brands the offer is about, a car brand on a car loan: `name`, `what`, and `logo` naming the `BADGE_n` reference of its logo; their logos may appear), an optional `direction`, the `mechanism` to rebuild for our brand, and in `/3`: `viewer.market` (`country`, `language`, `currency`), `brand_presence` (`standard` | `native`: a native disguise carries the logo as its sender and no big button), `recognition` (the kind of product a stranger must take it for at a glance, never the format), `native` (the ad is a screen people know, drawn exactly from HTML at every size by the free `static_native` node instead of painted: `x_post` (a post or a thread, with `reply_to`), `whatsapp`, `imessage`, `trustpilot`, `google_reviews`, `notes`, `slack`; on a 9:16 the app is drawn full screen as a phone shows it: status bar and app header pinned at the top, the reply bar and tab bar (or a keyboard under a chat or note) pinned at the bottom, the posts or messages as large as the clear band holds, and the frame's headline, caption and logo a card in the brand's colours under the app's header; a wall of reviews (`trustpilot`, `google_reviews`) takes up to 16 real reviews, the first two (`FEATURED_REVIEWS`) the ad's copy set whole in the safe area and the rest filling the whole frame behind them as large, sharp cards, turned and cut by its edges, because a wall wins on how many there are; a review longer than 160 characters is shown cut at a word with "…", and that cut text is the copy the review reads back; a chat (`whatsapp`, `imessage`) on a 9:16 never shrinks below about 62% of the phone scale: its earliest messages scroll up under the header, as the app does, the node's optional `scrolled` output lists the copy lines scrolled off that size, and the run review expects those lines nowhere on it; a logo file whose drawing runs off its own edge is set aside for a whole one, and a square mark where a wordmark goes is drawn larger; the run review never flags the app's own interface words (status bar, keyboard) as under Reels; the image model renders only the screen's `photos`, each a plain phone photo of ordinary people of the market; adding the block switches an existing ad on its next run, `--fix` does not apply, an invented brand handle or a rating the Knowledge does not hold (either decimal mark) is refused as `native_screen_invalid`, and a photo of ours goes in as its `image`: a native photo takes no `avatar`, since nothing would cast it), a role on each third-party brand (`partner` | `proof` | `category_signal`), a reference's `image` (wired as a picture of ours) and a person's `avatar` (the company avatar cast), and `concept` (`why_it_wins` in five layers, `load_bearing` elements with `have`/`adapt`/`missing` and `ours`, `verdict` `use`|`adapt`|`reject`, `transfer_note`; seeded from `baker winning-ads fit`). Every render keeps the load-bearing elements; `validate`/`run` refuse a `reject` or a `missing` element with nothing named (`concept_does_not_fit`) and a rating the Knowledge does not hold (`proof_without_source`); a line citing a review platform gets its real badge wired. **You never write a model prompt**: every render prompt is compiled from this file for the model it runs on (short labelled sections for gpt-image, sentences with reasons for Gemini; each line quoted once; pictures numbered with one role each; colours from the brand; realism as capture cues; the safe area from its one owner; constraints last), on every `validate` and `run`, and shown on its node. A `prompt.json` written before the ad spec (the describe blueprint) is read through an adapter and keeps working. The source's own read (`source`, and the library's `reference`) stays as bookkeeping and is never sent to a model.
 2. **element selection** (`~google/gemini-flash-latest`) — picks the **main, identity-critical** elements (the brand logo, a showcased product, a trust badge) **plus any foreground/hero person or animal** — the emotional focal point — even a generic one, because a free-generated face/muzzle reads as AI and grows artifacts; the emotional hero always gets a real-reference slot. When the advertiser's logo appears in **more than one lockup** (a square/icon **mark** and a horizontal **wordmark**), each is emitted as its **own** element (e.g. `LOGO_MARK`, `LOGO_WORDMARK`). The describe pass also records the ad's **typography** (each typeface's classification, a best-guess family, and its weight/case), which seeds the spec's `brand.typeface`. Background extras are dropped. Each element becomes a `references` entry in the spec, named by its role and naming the canvas slot it fills, so the file says which slot grounds which subject.
 
 It then scaffolds a canvas on which **the image model designs the finished ad**, holding only what `prompt.json` holds; no element is mandatory:
 
 - Up to **three variants** (`--takes`, default 3) render on the model the router picks for the concept (today `openai/gpt-image-2.5-sunburst` at quality `max`; `--gen-model` overrides), each a finished ad: the picture and every word of the copy. Each variant is **its own request** (tagged `static_variant_call`), run in parallel on its own direction (the scene as written; another picture of the same message, by the scene's look; another line leading, named by number so no word changes), and a `collect` tagged `static_variants` gathers them, on every model: one request with `n` costs the same (`n` is billed per picture) but gave fewer distinct options, so the graph never uses it. The brief quotes every line of `copy` verbatim and asks for each once; it is compiled again from `prompt.json` on every `validate` and `run`, so an edit re-keys exactly the renders it touches. The brand's **typeface** is shown to the model as a `font_specimen` set on the `designed` lines: a file in `src/brand/fonts/` (via `src/brand/tokens.json` when present), else the heading family `tokens.json` or `BRAND.md` names, fetched as a web font file; an `in_scene` line keeps its own lettering only on the real packaging, signage or screen its `on` names (handwriting on post-its, whiteboards or chalkboards, and a line naming no surface, are set as designed type). Each line's `role` sets its size: a price or stat (else the headline) leads, the `cta` is a large solid button, fine print is smallest. **The logo is optional**: the real file in `src/brand/logos/` goes to the model as a reference to reproduce once, at least a fifth of the frame wide, where it contrasts or on a small brand-colour plate, only when `prompt.json` has `"logo": true` (an older blueprint: an `advertiser_brand` entry in `brands_logos`) and the file exists; no logo is never an error, never a slot to fill and never a stand-in. Setting it false drops the logo on the next run; adding one to a canvas scaffolded without the file needs the file in `src/brand/logos/` and a re-scaffold. `prompt.json` also carries a **`brand`** block (ground, ink, accent, CTA and palette, from `src/brand/tokens.json` roles or `BRAND.md`); a colour outside it never reaches the model. The brief describes a complete ad at the master's shape and asks only for its edge margin (see below). `baker canvas set-prompt` refuses these nodes and names `prompt.json`.
-- A **`static_prefilter`** node (tagged `variant_prefilter`) drops variants that fail the deterministic checks before the judge sees them, never all of them; an `image_select` **judge** (tagged `pick`) compares the rest pairwise against the compiled brief (one variant passes through unjudged).
-- **Every placement format is made from the chosen picture, and no size is a crop or a paste of another.** The pick goes to exact pixels as the master (an `imagemagick` node tagged `static_master`) at the lead size: the 4:5 when a feed size is delivered, else the square, else the first portrait size, rendered at the model's nearest ratio (3:4 on GPT Image) and filled to Meta's pixels. It is delivered as it is. Every other size is **re-laid out** from it by one request (`image_generate` node `adapt_<size>`, tagged `static_adapt_format`, on the router's `relayout` model, GPT Image 2.5 at quality `max`): the master, the variants' own product and people pictures and the logo file as references and a brief compiled from `prompt.json` for that frame on every run (`static-adapt.ts`): the exact lines and their hierarchy, the same people (same face, same clothes), product and label, logo and none the chosen ad does not show, the words and the button re-scaled for the frame, the frame stated as zones inside its safe area (on a 9:16, the words set large through the Reels band, the headline across most of its width, lists restacked rather than shrunk, and the ad's own picture larger above and below it, never flat colour), never the master shrunk onto a background, set as a square or cut. A `size_<size>` node (`static_size`) brings it to Meta's exact pixels (9:16 at 1080×1920, 1:1 at 1080×1080; a landscape size the same way). A 9:16 that Reels or Stories covers then passes a free `static_safe_fit` node (`safe_9x16`), which shrinks and shifts the picture just enough (to 85% at most) to bring a word or the logo a few points past the band back inside it; a canvas scaffolded before it gets the node on its next run, every other node kept. The master shape is one decision in the code (`DEFAULT_MASTER_SHAPE`: `relayout`; the `9x16_band` and `4x5_adapt` shapes, kept for the bench, make the sizes by cutting or reframing the master, and the person rating them rejected every such way; a canvas from the retired band master, which extended a square master into the 9:16, is upgraded to the re-layout graph on its next run). Placing the logo file on every size instead of having the model draw it (`logo_composite` nodes) is a scaffold option, off until it is proven on the bench.
-- Readers go by these roles, never by node ids: the run review, `--fix`, `--use-variant`, `--regenerate static_variants`, the render budget and the dashboard. A static ad scaffolded before this graph (`ad_a`/`ad_b`/`ad_c`, `pick`, re-lays `ad_4x5`/`ad_9x16`, `fit_4x5`) is **upgraded in place** on its next `validate` or `run`: prompt.json, every asset slot, the model and the formats are kept, the render part becomes this graph, and the creative is not reported as changed by the upgrade. Its old runs stay readable.
-- Detected elements get **one `[TODO]` ingest slot** each (never the logo), labelled by role (`el_product_1`, `el_person_1`), never by the source's brand. The original image stays on the canvas as the record of the source and is **never given to the render**: the generator copies what it is shown, whatever the prompt says. Each **animal hero** is fused into a generated **multi-view reference sheet** (`image_reference_sheet`) that the render grounds on; pass `--skip-actor-sheets` to ground on the dropped photo. A **person** with no avatar has no slot: the render draws an invented, realistic person from the element's description. With `--avatar <handle>` (for an actor who recurs across ads) the slot ingests the Avatar's identity sheet and `prompt.json` describes that person in the Avatar's own words.
+- A **`static_prefilter`** node (tagged `variant_prefilter`) drops variants that fail the deterministic checks before the judge sees them, never all of them; its **judge** (`judge_model`, `judge_prompt`) compares the rest pairwise against the compiled brief when the checks tie (one variant passes through unjudged). A canvas scaffolded with a separate `image_select` judge (tagged `pick`) has it folded into the prefilter on its next run.
+- **Every placement format is made from the chosen picture, and no size is a crop or a paste of another.** The pick goes to exact pixels as the master (an `imagemagick` node tagged `static_master`) at the lead size: the 4:5 when a feed size is delivered, else the square, else the first portrait size, rendered at the model's nearest ratio (3:4 on GPT Image) and filled to Meta's pixels. It is delivered as it is. Every other size (9:16, 1:1, 16:9, …) is **re-laid out** from the master by one request (`image_generate` node `adapt_<size>`, tagged `static_adapt_format`, on the router's `relayout` model, GPT Image 2.5 at quality `max`): the master, the variants' own product and people pictures and the logo file as references and a brief compiled from `prompt.json` for that frame on every run (`static-adapt.ts`): the exact lines and their hierarchy, the same people (same face, same clothes), product and label, logo and none the chosen ad does not show, the words and the button re-scaled for the frame, the frame stated as zones inside its safe area, never the master shrunk onto a background, set as a square or cut. A `size_<size>` node (`static_size`) brings it to Meta's exact pixels (1:1 at 1080×1080; a landscape size the same way). Every 9:16 Reels or Stories covers then passes a free `static_safe_fit` node (`safe_<size>`), which shrinks and shifts the picture just enough (to 85% at most) to bring a word or the logo a few points past the band back inside it; a layout further out passes unchanged and the run lays it out again. The 9:16 is never extended instead (the 4:5 set whole in the clear band, its scene continued to the frame) nor set over its own blurred copy: rated in October 2026, the first showed the ad small inside a continued scene ("text super tiny") and the second "looks bad". The master shape is one decision in the code (`DEFAULT_MASTER_SHAPE`: `relayout`; the `9x16_band` and `4x5_adapt` shapes, kept for the bench, make the sizes by cutting or reframing the master, and the person rating them rejected every such way; a canvas from the retired band master, which extended a square master into the 9:16, is upgraded to the re-layout graph on its next run). Placing the logo file on every size instead of having the model draw it (`logo_composite` nodes) is a scaffold option, off until it is proven on the bench.
+- Readers go by these roles, never by node ids: the run review, `--fix`, `--use-variant`, `--regenerate static_variants`, and the dashboard. A static ad scaffolded before this graph (`ad_a`/`ad_b`/`ad_c`, `pick`, re-lays `ad_4x5`/`ad_9x16`, `fit_4x5`) is **upgraded in place** on its next `validate` or `run`: prompt.json, every asset slot, the model and the formats are kept, the render part becomes this graph, and the creative is not reported as changed by the upgrade. Its old runs stay readable.
+- Detected elements are **never slots to fill**: the source's people, products, animals and badges are its world. The original image stays on the canvas as the record of the source; when `prompt.json` keeps its concept (`concept.verdict` `use` or `adapt`) the first renders are shown it **for its structure only**, and otherwise it never reaches a render. Only the people cast from a company Avatar and the brand's logo become pictures on the canvas; everything else is described in words (a picture of ours is added through a reference's `image`). A **person** with no avatar needs no slot and no casting: the render draws an invented, realistic person from the element's description. With `--avatar <handle>` (for an actor who recurs across ads) the slot ingests the Avatar's identity sheet and `prompt.json` describes that person in the Avatar's own words. A reference's `avatar` names who was cast and draws no face on its own: its face reaches the render only with a picture behind it (the `--avatar` slot, or the identity sheet URL from `baker avatars get <handle>` as the reference's `image`).
 
 The canvas is validated before it is written. stdout returns `{ ok, canvas_path, prompt_path, models, aspect_ratio, placements, output_formats, stats, checklist }`; the **checklist** carries `summary` (what a run does: candidates, judge, type, logo, formats), `font`, `logo`, `in_scene_text`, `formats` and every real asset to drop in.
 
-**Placement formats.** The canvas delivers the platform's full placement set, each format at its exact size, collected under `final` with the format as its label. The set is derived from `--platform` (`meta` → 4:5 + 9:16, `google` → 1:1 + 4:5 + 16:9, `linkedin` → 1:1, `tiktok` → 9:16, `youtube` → 16:9); override it with `--placements <preset>` or `--placements none` for a single format. The report's `aspect_ratio` is the shape the variants really render at (the model's nearest to the master's format), and `format_plan` names the master and how each size is made from it (`master`, `crop` with its rectangle, or `adapt`).
+**Placement formats.** The canvas delivers every size where the ad runs, each at its exact size, collected under `final` with the format as its label. `--platforms meta,tiktok,google_display,…` (platforms or placement ids from the registry in `@baker/canvas-contract` `AD_PLACEMENTS`: meta, google_display, google_pmax, google_demandgen, google_search, youtube, linkedin, tiktok, x, reddit, pinterest, snapchat, chatgpt) delivers every size their placements require, each kept clear of its own placement's interface (a TikTok 9:16 of TikTok's buttons and caption), and lists fixed banners (`728x90`, …) in the report as `banners`, never made from the picture (a destination of banners alone is refused with `invalid_platforms` before anything bills). A 9:16 for another platform keeps clear of its interface even when `--target-adset` narrows Meta's placements. Without it the set is derived from `--platform` (`meta` → 4:5 + 9:16, `google` → 1:1 + 4:5 + 16:9, `linkedin` → 1:1, `tiktok` → 9:16, `youtube` → 16:9); override it with `--placements <preset>` or `--placements none` for a single format. The report's `aspect_ratio` is the shape the variants really render at (the model's nearest to the master's format), and `format_plan` names the master and how each size is made from it (`master`, `crop` with its rectangle, or `adapt`).
 
 **Who sees the ad.** `--target-adset <id>` names the Meta ad set the ad will run in. Before anything is billed, the viewer its targeting reaches is read exactly as `baker ads meta adsets --audience` reads it (`temperature`, what they already `knows`, `placements`) and written to `_definition.md` as `targetAdSet` and `viewer` (an existing definition keeps every other field). The sizes, the brief and the prefilter keep clear of that viewer's placements (`staticSafeArea`; `automatic`, a bare `facebook`/`instagram` or no placements is Advantage+, Reels included), and every `baker canvas run` checks the ad against that viewer. With no ad account, give the viewer from the Knowledge: `--viewer-temperature cold|warm|hot [--viewer-knows brand,offer,product] [--viewer-placements <placement,…>]`, recorded with `source: knowledge`; next to `--target-adset` they correct the read, and its `reasons` say so. A bad value exits 2 (`invalid_viewer`) before any billed pass; an ad set that cannot be read never stops the scaffold (a hint says why, and the sizes keep clear of every placement). With neither flag, a `viewer` (or `targetAdSet`) already in `_definition.md` is used as it is. stdout carries `target_adset`, `viewer` and a `checklist.viewer` line.
 
-**After a run** the CLI reads the finished ad back on what makes a static win, not a checklist of elements: every delivered image's pixel shape against its format; **every line of `prompt.json`'s copy (in-scene and designed alike) read back with its box by a box reader and compared** (a line more than 2% off, missing, or joined by words nobody asked for is blocking; case and punctuation fold, accents do not); **every word and the logo measured against the format's safe area** (`staticSafeArea`, the one owner of those areas: on a 9:16, the Reels and Stories interface unless the ad set's placements rule them out, which is blocking; on a feed image such as a 4:5, only a design margin, so words close to the edge are a warning and only words the frame cuts are blocking); a copy line written over two lines (`"15%\noff"`) is matched part by part wherever the reader finds each; designed type in a colour far from the brand palette is a warning. When the reader cannot answer, the vision pass transcribes the copy instead and judges the safe zones by eye. Then one vision pass for clipping, overlaps, artifacts, one dominant idea, the brand with the logo covered and the concept's mechanisms, with every format shown a second time at phone-feed thumbnail size (240px wide, same call): a lead message unreadable at that size is blocking. A missing logo, CTA or copy is never a finding; a logo that is drawn and broken is. stdout carries `verdict` (`ready` | `ready_with_notes` | `needs_fixes` | `not_reviewed`), `judge` (the per-candidate scores) and hints that end with what the run made; the findings are stored on the run as its `review`. `needs_fixes` means the ad is not done even though every node completed, and the run is recorded with `presentable: false` so it is never shown as a version; every fix is an edit to `prompt.json` and a re-run (wrong words: the lines in `copy`; wrong picture: the `scene`; a bad roll: `--regenerate static_variants`; one re-laid-out size wrong while the chosen picture is right: `--regenerate adapt_<size>`, named in the hints). The checks rank the variants before the pick and the judge only breaks an exact tie, comparing pairwise against the compiled brief, knowing the brand and the viewer; every variant stays on offer to the person in Baker. **Render budget:** a static ad gets 3 renders per round (`RENDERS_PER_ROUND`), and a round is one version the person sees; the backend counts them (`GET /api/canvas/render-budget`), with a local ledger only when it cannot be reached. The fourth exits 2 with `RENDER_BUDGET` without rendering. The person's reaction opens a new round: their answer in Baker, or `baker creatives feedback record` / `baker creatives select` from the chat. Resuming an interrupted run does not count again, and a finished run that billed nothing (an outage: timeout, network, a provider that says retry) gives its render back. When the round is spent and the verdict is still `needs_fixes`, the hints say to show the best version and name the defect in plain words rather than render again.
+**After a run** the CLI reads the finished ad back on what makes a static win, not a checklist of elements: every delivered image's pixel shape against its format; **every line of `prompt.json`'s copy (in-scene and designed alike) read back with its box by a box reader and compared** (a line more than 2% off, missing, or joined by words nobody asked for is blocking; case and punctuation fold, as does a capital I read as an l (one stroke in a sans-serif), accents do not; words the picture adds are blocking only when set at the copy's usual size or making a claim (a figure, a price, an offer), and otherwise a note, as a screen's code, a poster or a mug in the scene; a copy line is "on the ad twice" only when most of its words repeat, never a word or two echoed on a prop; on a native screen, which code typesets from the spec, a line read differently is the reader's error and only a missing one counts); **the ad's own copy and its logo measured against the format's safe area** (`staticSafeArea`, the one owner of those areas; `must-show.ts` decides what is measured: never what the scene prints, nor a copy line printed again in the scene while the line itself sits clear). On a 9:16, which runs in Stories and Reels alike, what Stories' header or bottom bar covers (and any other app's interface named) is said as "where Stories and Reels cover it" ("where Reels covers it" for an ad set that runs only in Reels) and blocks for the logo, the button, the headline, the figure and the offer, while a supporting line (body, caption, label, small print) there is a note; what only Reels' taller footer covers is one note, and only for the lead line, since Stories shows it and the 9:16 is laid out in Stories' band; words merely close to where an app draws are no finding. The steps that place a size clear of an interface (`static_safe_fit`, `static_place`) are told the same copy (`prompt.json`'s lines, their roles, and the brand's name and the logo's words, a wordmark read close to them counting as the logo) and move the ad for those alone, a supporting line included only where the frame cuts it; on a feed image such as a 4:5, only a design margin, so words close to the edge are a warning and only words the frame cuts are blocking); a copy line written over two lines (`"15%\noff"`) is matched part by part wherever the reader finds each; designed type in a colour far from the brand palette is a warning; supporting lines too small to read on a phone are one note per size; the main line set smaller than a supporting one is a note, the brand's wordmark never counted as part of a line that starts with its name; the logo set twice is a note, a far smaller mark printed on something in the scene (a hoodie, a bag) is not. When the copy names a real person ("Lautaro Suarez, CTO at Vibiz"), the run casts them as themselves before it renders (`baker avatars research --create`, their avatar's sheet added to `prompt.json` `references` as a `PERSON_n`), and a face shown as a named person the ad has no photo of is blocking. When the reader cannot answer, the vision pass transcribes the copy instead and judges the safe zones by eye. Then one vision pass for clipping, overlaps, artifacts, one dominant idea, the brand with the logo covered and the concept's mechanisms, with every format shown a second time at phone-feed thumbnail size (240px wide, same call): a lead message unreadable at that size is blocking. A missing logo, CTA or copy is never a finding; a logo that is drawn and broken is. stdout carries `verdict` (`ready` | `ready_with_notes` | `needs_fixes` | `not_reviewed`), `judge` (the per-candidate scores) and hints that end with what the run made; the findings are stored on the run as its `review`. `needs_fixes` means the ad is not done even though every node completed, and the run is recorded with `presentable: false` so it is never shown as a version; every fix is an edit to `prompt.json` and a re-run (wrong words: the lines in `copy`; wrong picture: the `scene`; a bad roll: `--regenerate static_variants`; one size wrong while the chosen picture is right: `--regenerate adapt_<size>`, which re-lays it out, named in the hints); a blocking defect is first corrected by the run itself (`--autofix`, above). The checks rank the variants before the pick and the judge only breaks an exact tie, comparing pairwise against the compiled brief, knowing the brand and the viewer; every variant stays on offer to the person in Baker. There is no render limit. When a defect still stands after the automatic rounds, the hints say to fix it with a `--fix`, then by changing the spec or the approach, and if it still cannot be fixed to show the best version and name what is left in one plain sentence, without a question.
 
 The render prompt is **model-aware**: each model's profile (`src/engine/lib/prompt-profiles.ts`) sets its shape, word budget, picture order and request params. gpt-image gets short labelled sections with the constraints last; Gemini gets sentences that give the reason and a positive closing list of the only words allowed. A profile change, or a new model on the router (`src/engine/lib/image-model-router.ts`), is re-run on the static bench (`scripts/creative-evals`, `--job static`) before it ships. Run `baker canvas critique` on the scaffolded canvas for an advisory read before you spend.
 
@@ -5976,9 +6036,9 @@ baker canvas run ./static-ad.canvas.json
 | `--gen-model <id>` | the router's pick for the concept | Override the render model. An unknown model id in any `--*-model` flag fails the command (`unknown_model`) before anything is billed. |
 | `--takes <n>` | `3` | Variants of the finished ad rendered (one request each) and judged (1-3); `1` skips the judge. |
 | `--platform <channel>` | `meta` | `_definition.md` channel **and** the default placement set (`meta`\|`google`\|`linkedin`\|`tiktok`\|`youtube`\|`x`\|`other`; `x`/`other` deliver one format). |
-| `--placements <preset\|none>` | derived from `--platform` | Override the set: `meta_feed`\|`meta_feed_stories`\|`meta_all`\|`linkedin`\|`google_demandgen`\|`tiktok`\|`youtube`, or `none` for a single base ad. |
+| `--platforms <list>` | none | Where the ad runs, comma-separated platforms or placement ids: every size they require, each kept clear of its own interface. Wins over `--placements`. Banners alone are refused. |
+| `--placements <preset\|none>` | derived from `--platform` | An older fixed set: `meta_feed`\|`meta_feed_stories`\|`meta_all`\|`linkedin`\|`google_demandgen`\|`tiktok`\|`youtube`, or `none` for a single base ad. |
 | `--aspect <ratio>` | inferred from the image, else `9:16` | Force the lead format (it leads the placement set when the set has it). |
-| `--skip-actor-sheets` | off | Ground each animal hero on its lone dropped photo instead of a generated multi-view reference sheet. |
 | `--avatar <handles>` | — | Cast company avatars as the image's people, comma-separated in the order they were detected (`marta` or `marta,joan`). Each person slot grounds on that avatar's identity sheet and takes its subject description. Checked before any billed pass; an unknown, not-ready or sheetless avatar fails with code `avatar`. A handle with no person to play is reported in `hints` as `NOT CAST`. |
 
 Scaffolding runs (and bills) the two vision passes; **running** the result generates billed images. `baker canvas validate` does not check that the `[TODO]` paths exist — supply the real files before `run`.