@eigenpal/cli 0.10.44 → 0.10.46
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +6 -0
- package/dist/cli.js +542 -2872
- package/dist/skill/reference/step-types.md +10 -3
- package/package.json +1 -1
|
@@ -365,8 +365,10 @@ Extract text from documents (PDF, DOCX, images) using OCR or vision models
|
|
|
365
365
|
| Field | Type | Required | Default | Description |
|
|
366
366
|
| --- | --- | --- | --- | --- |
|
|
367
367
|
| `input` | string | yes | | Storage reference or template expression for the document |
|
|
368
|
+
| `parseMode` | `"ocr"` \| `"vision"` | no | | Base parser for PDF/image inputs. OCR is the default; vision uses the selected LLM. Text and Office files always use native parsing. |
|
|
368
369
|
| `ocrModel` | string | no | | OCR provider ID for PDF/image parsing |
|
|
369
370
|
| `llmModel` | string | no | | LLM provider ID for vision-based parsing |
|
|
371
|
+
| `figureModel` | string | no | | Vision model used only for the optional figure-description pass |
|
|
370
372
|
| `maxConcurrency` | number | no | `3` | Max concurrent VLM batch requests |
|
|
371
373
|
| `pagesPerBatch` | number | no | `5` | Number of page images per VLM request |
|
|
372
374
|
| `pdfRenderScale` | number | no | `1` | Scale factor for rendering PDF pages before VLM parsing. Higher values produce sharper images at larger payload sizes. |
|
|
@@ -382,9 +384,14 @@ Extract text from documents (PDF, DOCX, images) using OCR or vision models
|
|
|
382
384
|
|
|
383
385
|
| Field | Type | Required | Default | Description |
|
|
384
386
|
| --- | --- | --- | --- | --- |
|
|
385
|
-
| `
|
|
386
|
-
| `
|
|
387
|
-
| `
|
|
387
|
+
| `document` | object | yes | | |
|
|
388
|
+
| `usage` | object | yes | | |
|
|
389
|
+
| `pages` | array<object> | yes | | |
|
|
390
|
+
| `text` | string | yes | | Combined text from all pages |
|
|
391
|
+
| `parserType` | `"plaintext"` \| `"office"` \| `"llm-vision"` \| `"ocr"` | yes | | |
|
|
392
|
+
| `parserVersion` | string | no | | |
|
|
393
|
+
| `model` | string | no | | Model used (for LLM/OCR parsers) |
|
|
394
|
+
| `structured` | record<string, unknown> | no | | Canonical structured document with ordered blocks, regions, bounding boxes, tables, figures, and chunks when supported by the parser |
|
|
388
395
|
|
|
389
396
|
#### `ai.extract` — Extract Data
|
|
390
397
|
|