Manifest Response Format (V1 to V2)

This guide covers every field-level change in the /v2/generate-manifest response. For request format migration (how to call the API), see the Manifest Migration guide.

Overview

The V2 manifest response is significantly richer than V1. The changes fall into these categories:

data-variant=warning

data-slots=text

Document-level fields

CHANGED — The old manifest had a document sub-object with different field names. The new format has a richer top-level document object.

Field renames

New document fields

Layer type taxonomy

data-variant=warning

data-slots=text

CHANGED — Type names renamed to snake_case; artboards promoted to a distinct type.

In the old format, artboards were layerSection nodes indistinguishable from regular groups. The new format promotes artboards to type: "artboard" with dedicated typeAttributes.artboard data.

Per-layer common fields

bounds — CHANGED format

data-variant=warning

data-slots=text

Old format uses {height, left, top, width}:

{ "height": 200, "left": 10, "top": 20, "width": 300 }

New format uses {left, top, right, bottom} (no height/width):

{ "left": 10, "top": 20, "right": 310, "bottom": 220 }

thumbnail — CHANGED (restructured)

data-variant=warning

data-slots=text

Old format: plain string (pre-signed S3 URL):

"thumbnail": "https://s3.amazonaws.com/..."

New format: object with media type and short-URL:

"thumbnail": {
  "mediaType": "image/jpeg",
  "url": "https://photoshop-api.adobe.io/v2/short-url/urn:aaid:ps:..."
}

The thumbnail is optional — only present when the caller requests thumbnails. The URL is now a /v2/short-url/ redirect rather than a direct S3 URL.

Children nesting key — CHANGED

• Old: all nested layers use children[] at every level
• New: groups use children[]; artboards use layers[] for their children (see Artboard structure)

blendOptions — CHANGED (extended)

Old blendOptions only carried {blendMode, opacity}.

New blendOptions adds:

pixelMask — CHANGED (was mask)

Old format uses a mask field at the layer level:

"mask": {
  "enabled": true,
  "linked": true,
  "offset": { "x": 0, "y": 0 },
  "type": "layer"
}

New format renames and restructures this as pixelMask (can be null or populated):

"pixelMask": {
  "hasMask": true,
  "extendOpaque": true,
  "bounds": { "left": 0, "top": 0, "right": 2100, "bottom": 1500 },
  "enabled": true,
  "linked": true,
  "offset": { "horizontal": 0, "vertical": 0 }
}

Fields removed in V2

New fields in V2 only

Adjustment layers

Type rename: adjustmentLayer → adjustment_layer

adjustments block — CHANGED (restructured with type discriminant)

Old format uses a camelCase sub-object key per adjustment type:

{ "adjustments": { "vibrance": { "saturation": -24, "vibrance": 15 } } }
{ "adjustments": { "brightnessContrast": { "brightness": 18, "contrast": 36 } } }

New format adds a type discriminant (snake_case) and restructures some sub-fields:

{ "adjustments": { "type": "vibrance", "vibrance": 15, "saturation": -24 } }
{ "adjustments": { "type": "brightness_contrast", "brightnessContrast": { "brightness": 18, "contrast": 36 } } }

Per-type change summary

Background layer

Type rename: backgroundLayer → background_layer

The background layer in the new format carries additional sub-fields not present in the old format. Background layer entries do not include typeAttributes (they match other layer types such as text, group, and adjustment).

Example V2 background layer:

{
  "type": "background_layer",
  "id": 1,
  "name": "Background",
  "backgroundLayer": true,
  "background": {
    "canConvertToLayer": true
  }
}

Fill / solid color layer

Type rename: fillLayer → solid_color_layer

fill.solidColor.rgb — unchanged

Both old and new use identical structure and value range (8-bit, 0–255):

{ "red": 255, "green": 0, "blue": 0 }

Vector mask — NEW in V2

Old fillLayer had no vector mask information. New solid_color_layer exposes:

{
  "vectorMask": {
    "hasMask": true,
    "bounds": { "left": 0, "top": 0, "right": 500, "bottom": 500 }
  }
}

layerSettings.vectorMask may also carry offset sub-data for the mask.

Artboard structure

data-variant=warning

data-slots=text

Old format — no distinct artboard type

Artboards were plain layerSection nodes, indistinguishable from regular groups:

{
  "type": "layerSection",
  "id": 1,
  "name": "iPhone 16 Pro – 1",
  "children": [ ... ]
}

New format — distinct artboard type with rich metadata

{
  "type": "artboard",
  "id": 1,
  "name": "iPhone 16 Pro – 1",
  "typeAttributes": {
    "artboard": {
      "type": "artboard",
      "artboardBackgroundType": 1,
      "artboardPresetName": "iPhone 16 Pro",
      "artboardRect": {
        "type": "classFloatRect",
        "left": 0,
        "top": 0,
        "right": 393,
        "bottom": 852
      },
      "color": {
        "type": "RGBColor",
        "red": 255,
        "grain": 255,
        "blue": 255
      },
      "guideIDs": []
    },
    "layerSection": "layerSectionStart",
    "type": "layerSection"
  },
  "layers": [ ... ]
}

New artboard-specific fields

Children key difference

Smart object detail

Old smartObject block

{
  "instanceId": "16da7404-...",
  "linked": false,
  "name": "foo.psd",
  "path": "foo.psd",
  "type": "image/vnd.adobe.photoshop"
}

New smart object block (heavily restructured)

{
  "isSmartObject": true,
  "isLinked": false,
  "isValid": true,
  "smartObjectData": {
    "type": "smartObject",
    "transform": [x0, y0, x1, y1, x2, y2, x3, y3],
    "fileInfo": {
      "name": "foo.psd",
      "path": "foo.psd",
      "fileType": "ps3",
      "linked": false
    }
  },
  "extracted": {
    "mediaType": "application/octet-stream",
    "url": "https://photoshop-api.adobe.io/v2/short-url/urn:..."
  }
}

extracted is only present for non-linked (embedded) smart objects. ccLibrariesElement inside fileInfo is only present for CC Library-linked assets.

Field-by-field comparison

Text layer detail

Character styles — CHANGED and NEW

Old characterStyles fields:

{
  "fontAvailable": true,
  "fontColor": { "rgb": { "red": 0, "green": 0, "blue": 0 } },
  "fontName": "PostScriptName",
  "fontSize": 12,
  "orientation": "horizontal"
}

New characterStyles fields:

orientation (old) is no longer a per-character-style field; it is now textOrientation at the text object level.

Paragraph styles — CHANGED and NEW

Old array was paragraphStyles:

[{ "alignment": "left" }]

New array is paragraphAttributes (renamed):

Text object top-level — CHANGED and NEW

Text frame — CHANGED

Old text frame fields:

{
  "type": "area",
  "frameType": "inPath"
}

New frame object:

{
  "type": "area",
  "bounds": { "top": 0, "left": 0, "bottom": 100, "right": 300 },
  "boundsRaw": { "left": 0, "top": 0, "right": 300, "bottom": 100 },
  "matrix": { "f_xx": 1, "f_xy": 0, "f_yx": 0, "f_yy": 1, "f_tx": 0, "f_ty": 0 },
  "baselineAlignment": { "flag": 0, "minimum": 0 }
}

Summary: new vs. changed

Fully new fields (absent in V1)

• Document-level metadata: bounds, imageMode, depth, iccProfile, xmp, title, resolution, globalLightingAngle, globalLightingAltitude, layerFXVisible, pixelScaleFactor
• protection array (replaces locked)
• userMask, vectorMask per-layer; pixelMask (restructured from mask)
• compositeFrame, referenceFrame, compositeDimensions per-layer bounds
• proportionalScaling, effects, layerSettings block
• Extended blendOptions fields: clipped, interior, fillOpacity, knockout, channelRestrictions, blendRanges
• Text: missingFonts, antiAliasing, textOrientation, frame.bounds, frame.boundsRaw, frame.matrix, frame.baselineAlignment
• Text warp: layerSettings.layerSpecific.textLayerSettings block
• Character style: fontAlpha, syntheticBold, syntheticItalic, underline, capsOption, lineHeight, letterSpacing, position
• Paragraph: writingDirection, firstLineIndent, startIndent, endIndent, spaceBefore, spaceAfter, position
• Smart object: isSmartObject, isValid, smartObjectData.transform, ccLibrariesElement, extracted
• Artboard: typeAttributes.artboard block, artboardPresetName, artboardRect, background color, guideIDs
• Background layer: backgroundLayer flag, background.canConvertToLayer (no typeAttributes)
• Adjustment adjustments.type discriminant; new adjustment types: curves, gradient_map_custom_stops
• version top-level field

Changed fields (present in both, different shape)

• Document field renames: bitDepth→depth, iccProfileName→iccProfile, name→title, height+width→bounds; photoshopBuild removed
• Layer type names (see Layer type taxonomy)
• bounds format: {height,left,top,width} → {left,top,right,bottom}
• locked (boolean) → protection (array of strings)
• rotate → rotation (renamed)
• mask → pixelMask (restructured; offset.{x,y} → offset.{horizontal,vertical}; extendOpaque and bounds added)
• thumbnail: plain string URL → {mediaType, url} object
• Artboard children key: children[] → layers[]
• Text font restructured: fontName/fontAvailable → font.postScriptName/font.isAvailable
• Text paragraph array: paragraphStyles → paragraphAttributes
• Text frame: {type, frameType} → frame object with bounds + matrix
• Smart object: flat {instanceId, linked, name, path, type} → nested {isLinked, isValid, smartObjectData: {fileInfo, transform}, extracted}
• Smart object MIME string type → fileType short code (e.g. "ps3")
• Adjustment adjustments block: per-type keys renamed to snake_case; type discriminant added; vibrance fields hoisted; exposure value renamed to exposureValue; hueSaturation.channels[] → hueSaturationAdjustments[]