@proveanything/smartlinks 2.0.32 → 2.0.34

package/dist/ai-tools.d.ts

@@ -23,6 +23,7 @@ export declare const AI_TOOL_NAMES: {
     readonly IMAGE_FROM_REFERENCE: "image.fromReference";
     readonly IMAGE_SEARCH_STOCK: "image.searchStock";
     readonly IMAGE_TRANSFORM: "image.transform";
+    readonly IMAGE_OCR: "image.ocr";
     readonly PDF_CREATE: "pdf.create";
     readonly PDF_FILL: "pdf.fill";
     readonly PDF_MERGE: "pdf.merge";

package/dist/ai-tools.js

@@ -20,6 +20,7 @@ export const AI_TOOL_NAMES = {
     IMAGE_FROM_REFERENCE: 'image.fromReference',
     IMAGE_SEARCH_STOCK: 'image.searchStock',
     IMAGE_TRANSFORM: 'image.transform',
+    IMAGE_OCR: 'image.ocr',
     PDF_CREATE: 'pdf.create',
     PDF_FILL: 'pdf.fill',
     PDF_MERGE: 'pdf.merge',
@@ -66,7 +67,9 @@ export const BUILTIN_AI_TOOLS = [
     { name: 'pdf.inspect', group: 'media', capabilities: ['web:read'],
         description: 'Cheaply introspect a PDF before any AI call: page count/sizes, whether each page has an extractable text layer, whether a raster image is present, and a routing recommendation (text-native → cheap text pass; curve-only/raster → vision/OCR). Deterministic, no AI.' },
     { name: 'pdf.render', group: 'media', capabilities: ['web:read'],
-        description: 'Rasterize a single PDF page to a PNG at a controllable DPI (e.g. 300 for small print) and return a hosted image URL — feed to a vision call or screenshot a page.' },
+        description: 'Rasterize a single PDF page — or just a region (clip {x0,y0,x1,y1}, 0–1 from top-left) — to a PNG at a controllable DPI and return a permanent hosted image URL. Max 40 MP per call; larger requests return code "too_large" with suggestedDpi.' },
+    { name: 'image.ocr', group: 'media', capabilities: ['web:read'],
+        description: 'Deterministic OCR (Google Vision, no generative model): exact characters with per-word confidence and pixel boxes, plus lines as printed. Input an imageUrl, or a PDF url + page + dpi + optional clip. For small print and outlined text — never "corrects" misspellings.' },
     { name: 'pdf.extract', group: 'media', capabilities: ['web:read', 'ai:vision'],
         description: 'Extract typed JSON from a PDF in one call given a schema and/or prompt. Auto-routes: text-layer pages use cheap text extraction; curve-only/raster pages are rendered and read with vision. Optional per-field confidence/source (includeConfidence) and bounding boxes (includeBoxes).' },
     { name: 'pdf.decodeBarcodes', group: 'media', capabilities: ['web:read'],

package/dist/docs/API_SUMMARY.md

@@ -1,6 +1,6 @@
 # Smartlinks API Summary
 
-Version: 2.0.32  |  Generated: 2026-10-01T17:18:44.658Z
+Version: 2.0.34  |  Generated: 2026-10-02T16:06:32.154Z
 
 This is a concise summary of all available API functions and types.
 
@@ -1447,10 +1447,32 @@ interface ImageTransformArgs {
 }
 ```
 
+**ImageOcrArgs** (interface)
+```typescript
+interface ImageOcrArgs {
+  imageUrl?: string
+  url?: string
+  page?: number
+  dpi?: number
+  clip?: PdfClip
+  languages?: string[]
+}
+```
+
+**PdfClip** (interface)
+```typescript
+interface PdfClip {
+  x0: number; y0: number; x1: number; y1: number
+}
+```
+
 **PdfCreateArgs** (interface)
 ```typescript
 interface PdfCreateArgs {
-  html: string; format?: string; landscape?: boolean
+  html: string
+  format?: string
+  landscape?: boolean
+  margin?: string
 }
 ```
 
@@ -1478,7 +1500,10 @@ interface PdfInspectArgs {
 **PdfRenderArgs** (interface)
 ```typescript
 interface PdfRenderArgs {
-  url: string; page?: number; dpi?: number
+  url: string
+  page?: number
+  dpi?: number
+  clip?: PdfClip
 }
 ```
 
@@ -1532,6 +1557,7 @@ interface AiToolArgsMap {
   'image.fromReference': ImageFromReferenceArgs
   'image.searchStock': ImageSearchStockArgs
   'image.transform': ImageTransformArgs
+  'image.ocr': ImageOcrArgs
   'pdf.create': PdfCreateArgs
   'pdf.fill': PdfFillArgs
   'pdf.merge': PdfMergeArgs
@@ -1590,7 +1616,7 @@ interface ImageDescribeResult {
 **HostedAssetResult** (interface)
 ```typescript
 interface HostedAssetResult {
-  hostedUrl: string | null; contentType?: string; info?: { width?: number; height?: number; format?: string; size?: number }
+
 }
 ```
 
@@ -1625,7 +1651,51 @@ interface PdfInspectResult {
 **PdfRenderResult** (interface)
 ```typescript
 interface PdfRenderResult {
-  url: string | null; page: number; pageCount: number; dpi: number; width: number; height: number
+  url: string | null; page: number; pageCount: number; dpi: number; width: number; height: number; clip?: PdfClip
+}
+```
+
+**PdfRenderTooLarge** (interface)
+```typescript
+interface PdfRenderTooLarge {
+  error: string; code: 'too_large'; requestedPixels: number; maxPixels: number; suggestedDpi: number; width: number; height: number
+}
+```
+
+**OcrBox** (interface)
+```typescript
+interface OcrBox {
+  x: number; y: number; w: number; h: number
+}
+```
+
+**OcrWord** (interface)
+```typescript
+interface OcrWord {
+  text: string; confidence: number; bbox: OcrBox
+}
+```
+
+**OcrLine** (interface)
+```typescript
+interface OcrLine {
+  text: string; confidence: number; bbox: OcrBox
+}
+```
+
+**ImageOcrResult** (interface)
+```typescript
+interface ImageOcrResult {
+  words: OcrWord[]
+  lines: OcrLine[]
+  text: string
+  width: number | null
+  height: number | null
+  languages: string[]
+  page?: number
+  dpi?: number
+  clip?: PdfClip
+  engine: string
 }
 ```
 

package/dist/docs/ai.md

@@ -327,7 +327,8 @@ in `input`.
 | `pdf.fill` | Fill an AcroForm PDF's fields (`{ field: value }`) → hosted URL. |
 | `pdf.merge` | Merge several PDFs into one, in order → hosted URL. |
 | `pdf.inspect` | Cheap, no-AI introspection: page count/sizes, which pages have a real text layer, raster present, and a routing hint (`text` vs `vision`). |
-| `pdf.render` | Rasterize one page to a PNG at a chosen DPI → hosted image URL (feed to vision, or screenshot a page). |
+| `pdf.render` | Rasterize one page — or just a region (`clip: { x0, y0, x1, y1 }`, 0–1 from top-left) — to a PNG at a chosen DPI → hosted image URL (permanent). Max 40 MP per call; larger requests return `code: "too_large"` + `suggestedDpi`. |
+| `image.ocr` | **Deterministic OCR** (Google Vision, no generative model): exact characters with per-word `confidence` (0–1) and pixel `bbox`, plus lines with spacing as printed. Input: `imageUrl`, or a PDF `url` + `page` + `dpi` + optional `clip`. Optional `languages` hints. Use for small print and text outlined to curves. |
 | `pdf.extract` | PDF → typed JSON in one call (schema and/or prompt). Auto-routes text vs vision. Optional per-field `confidence`/`source` (`includeConfidence`) and `bbox` (`includeBoxes`) in `fieldsMeta`. |
 | `pdf.decodeBarcodes` | Deterministically decode barcodes/QR on a page (WASM, no AI) → value + symbology + page + bbox + confidence. Use this for barcode digits, never vision. |
 | `pdf.inspectGraphics` | Prepress inspection: per-page path/image/outlined-text counts + colour spaces, plus named SPOT colours (e.g. "PANTONE 871 C"). No AI. |
@@ -341,18 +342,36 @@ in `input`.
   returns typed JSON). It routes itself, but you can drive the route yourself: call `pdf.inspect`
   first (deterministic, no AI) to see whether each page has a real text layer, then `pdf.extract`
   (cheap text path) or `pdf.render` → `image.describe`/vision for curve-only or raster artwork.
-- **Zoom in on small print** (INCI/allergen lists) → `pdf.render` at a high DPI (e.g. 300), then read
-  the PNG with vision.
+- **Read small print exactly** (INCI/allergen lists, net weight, text outlined to curves) → `image.ocr`
+  with a PDF `url`, `page`, `dpi: 300–400` and a `clip` around the panel. It returns the characters it
+  sees with a per-word confidence — it never "corrects" a misspelling the way a vision model can, so
+  flag words below ~0.9 for human review rather than re-reading them with a model.
+- **Zoom in visually** (layout, artwork) → `pdf.render` with `clip` to rasterise just a region at a high
+  DPI, instead of the whole sheet. Renders over 40 MP return `code: "too_large"` with `suggestedDpi`
+  — retry at that dpi or with a smaller clip, don't blind-retry.
 - **Barcodes / QR** → `pdf.decodeBarcodes` (deterministic WASM decode) — never trust vision for barcode
   digits; it hallucinates them.
 - **Prepress / print QA** (spot colours, colour spaces, vector vs raster) → `pdf.inspectGraphics`.
 - **A review UI that flags guessed fields** → `pdf.extract` with `includeConfidence` (per-field
   confidence + source) and `includeBoxes` (per-field bbox on text-native pages) in `fieldsMeta`.
 - **Produce a PDF** → `pdf.create` (HTML → PDF), `pdf.fill` (populate an AcroForm's fields),
-  `pdf.merge` (combine several). These return a hosted `hostedUrl`.
+  `pdf.merge` (combine several). These return a hosted `hostedUrl` (`pdf.create` also returns it as
+  `url`). `pdf.create` loads remote `<img>` URLs before rendering and honours `page-break-inside`; pass
+  `margin: "0"` when your HTML sets its own body margin (default page margin is 18mm/14mm).
 
 Every tool is also directly callable without the model loop via `ai.tools.run(collectionId, name,
-args)` — e.g. render a page or extract fields straight from a UI, no agent round-trip.
+args)` — e.g. render a page or extract fields straight from a UI, no agent round-trip:
+
+```ts
+// Read the ingredients panel of a sleeve exactly (deterministic OCR of a clipped region)
+const { result } = await SL.ai.tools.run(collectionId, 'image.ocr', {
+  url: sleevePdfUrl, page: 1, dpi: 400,
+  clip: { x0: 0.40, y0: 0.20, x1: 0.60, y1: 0.40 },
+  languages: ['en', 'fr'],
+})
+// result.lines → [{ text: 'INGREDIENTS: Aqua, Glycerin, …', confidence: 0.97, bbox: {…} }]
+const toReview = result.words.filter((w) => w.confidence < 0.9)
+```
 
 Discover tools two ways:
 - **Design time (typed):** import `BUILTIN_AI_TOOLS`, `AI_TOOL_NAMES`, and the per-tool arg types

package/dist/docs/app-objects.md

@@ -747,6 +747,39 @@ const results = await app.records.list(collectionId, appId, {
 
 > `q` is not a full-text index and does not return ranked results. For ranked relevance search over large corpora, use the Elasticsearch integration.
 
+**Admin: search other data fields (`qFields`).** On the admin endpoint, `qFields` chooses which top-level `data` keys `q` searches (comma-separated, max 5; default `label`):
+
+```typescript
+const page = await app.records.list(collectionId, appId, {
+  recordType: 'packaging-field',
+  q: 'best before',
+  qFields: 'value,label',       // match data.value OR data.label
+  limit: 100, offset: 0,
+}, true);
+```
+
+### Filtering on data fields (admin)
+
+On the admin endpoint, `dataFilter` filters on **top-level `data` keys** — an exact match, an `in:a,b,c` list, or an array (any value matches). Up to 10 keys; combine freely with `recordType`, `status`, `q`/`qFields` and pagination:
+
+```typescript
+const page = await app.records.list(collectionId, appId, {
+  recordType: 'packaging-field',
+  dataFilter: {
+    versionId: 'ver_123',               // data.versionId = 'ver_123'
+    zone: 'in:on_pack,unclear',         // data.zone is one of these
+  },
+  limit: 100, offset: 0,
+}, true);
+
+page.pagination.total    // matches across all pages
+page.pagination.hasMore  // keep paging while true
+```
+
+- Values are compared as text (`data->>key`); nested keys and joins aren't supported.
+- `limit` is capped at **100** per page — page with `offset` while `hasMore` is true.
+- `status` also accepts `in:a,b` (e.g. `status: 'in:active,draft'`).
+
 ### External ID / ETL Workflow
 
 `customId` and `sourceSystem` provide a stable external key pair for loading records from external systems (CMS, ERP, PIM, etc.):