@shop-os/modelkit 0.2.19 → 0.2.21

package/dist/catalog.js

@@ -162,7 +162,9 @@ function geminiImage(id, label, aliases, sizes, display, exposure, references) {
   const constraints = {
     aspectRatios: [...GEMINI_RATIOS],
     sizes,
-    supports: ["safetySettings", "temperature"]
+    supports: ["safetySettings", "temperature"],
+    // google-image sends MediaRef.label as a text part beside each reference.
+    prompt: { referenceSyntax: "caption" }
   };
   if (references !== void 0) constraints.references = references;
   const model = {
@@ -449,6 +451,7 @@ var IMAGE_MODELS = [
     constraints: {
       aspectRatios: ["1:1", "2:3", "3:2", "16:9", "9:16", "4:3", "3:4", "4:5", "5:4", "21:9"],
       sizes: ["1K"],
+      references: { images: 8 },
       // flux-2-pro's job type always returns image/png; only the seed is forwarded.
       supports: ["seed"]
     },
@@ -1244,12 +1247,14 @@ var TEXT_MODELS = [
 var VIDEO_USAGE = { unit: "seconds" };
 var SEEDANCE_RATIOS = ["21:9", "16:9", "4:3", "1:1", "3:4", "9:16", "adaptive"];
 var KLING_RATIOS = ["16:9", "9:16", "1:1", "adaptive"];
+var KLING_OMNI_RATIOS = ["16:9", "9:16", "1:1"];
 var OMNI_RATIOS = ["16:9", "9:16"];
 var GROK_RATIOS = ["1:1", "16:9", "9:16", "4:3", "3:4", "3:2", "2:3"];
 var H3_RATIOS = ["21:9", "16:9", "4:3", "1:1", "3:4", "9:16"];
 var H3_REFERENCE_RATIOS = ["adaptive", ...H3_RATIOS];
 var NEGATIVE_PROMPT = "blur, distort, and low quality";
 var FRAMES_EXCLUDE_REFERENCES = { framesExcludeReferences: true };
+var KLING_PROMPT = { referenceSyntax: "numbered", maxChars: 2500 };
 var BYTEDANCE_LOGO = "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/3bba5d81-4e57-4e71-911b-cdfa05acca2c_compressed.png";
 var GOOGLE_LOGO2 = "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/192485d2-a76a-4657-9c06-930b2c70badd_compressed.png";
 var VIDEO_MODELS = [
@@ -1270,9 +1275,14 @@ var VIDEO_MODELS = [
     constraints: {
       aspectRatios: [...OMNI_RATIOS],
       resolutions: ["720p", "1080p", "4k"],
+      // Veo also takes 4 and 6, but must be 8 at 1080p/4k or with reference images;
+      // [8] until the kit has a rule that forces it there (CHANGELOG 0.2.21).
       durations: [8],
       references: { images: 3 },
-      supports: ["lastFrame", "seed"]
+      supports: ["lastFrame", "seed"],
+      // 1,024 text tokens; 3,000 characters assumes ~2.9 characters per token.
+      prompt: { referenceSyntax: "numbered", maxChars: 3e3 },
+      audio: "always"
     },
     defaults: { durationSeconds: 8, resolution: "720p", aspectRatio: "16:9" },
     usage: { ...VIDEO_USAGE, slug: "veo" },
@@ -1282,7 +1292,8 @@ var VIDEO_MODELS = [
       description: "Google's highest-quality video generation model. Supports up to 4K resolution and 8-second clips with reference image conditioning.",
       logo: GOOGLE_LOGO2
     },
-    capabilities: { ...FRAMES_EXCLUDE_REFERENCES }
+    // Veo's lastFrame "must be used in combination with the image parameter".
+    capabilities: { ...FRAMES_EXCLUDE_REFERENCES, lastFrameRequiresFirstFrame: true }
   },
   {
     id: "veo-3.1-fast",
@@ -1302,7 +1313,8 @@ var VIDEO_MODELS = [
       resolutions: ["720p", "1080p", "4k"],
       durations: [4, 6, 8],
       references: { images: 3 },
-      supports: ["lastFrame", "seed"]
+      supports: ["lastFrame", "seed"],
+      audio: "always"
     },
     defaults: { durationSeconds: 8, resolution: "720p", aspectRatio: "16:9" },
     usage: { ...VIDEO_USAGE },
@@ -1330,7 +1342,8 @@ var VIDEO_MODELS = [
       aspectRatios: [...OMNI_RATIOS],
       resolutions: ["720p", "1080p"],
       durations: [4, 6, 8],
-      supports: ["lastFrame", "seed"]
+      supports: ["lastFrame", "seed"],
+      audio: "always"
     },
     defaults: { durationSeconds: 8, resolution: "720p", aspectRatio: "16:9" },
     usage: { ...VIDEO_USAGE },
@@ -1364,7 +1377,10 @@ var VIDEO_MODELS = [
       durations: [3, 4, 5, 6, 7, 8, 9, 10],
       references: { images: 6, videos: 1, videoSeconds: { max: 10 } },
       sourceVideoSeconds: { max: 10 },
-      supports: []
+      supports: [],
+      // generateOmniVideo sends MediaRef.label as a text part beside each image.
+      prompt: { referenceSyntax: "caption" },
+      audio: "always"
     },
     defaults: {},
     usage: { ...VIDEO_USAGE, slug: "gemini_omni" },
@@ -1397,8 +1413,12 @@ var VIDEO_MODELS = [
       // extension can upload a <= 10 s source with task "extend"; later ones
       // chain by `previousInteractionId` (the API rejects the id together
       // with a video task, so those calls carry the prompt alone).
-      supports: ["videoExtension"],
-      extension: { segmentSeconds: 10, maxTotalSeconds: 40 }
+      // First-to-last-frame interpolation: `<LAST_FRAME>` goes with `<FIRST_FRAME>`.
+      supports: ["videoExtension", "lastFrame"],
+      extension: { segmentSeconds: 10, maxTotalSeconds: 40 },
+      // generateOmniVideo sends MediaRef.label as a text part beside each image.
+      prompt: { referenceSyntax: "caption" },
+      audio: "always"
     },
     defaults: {},
     usage: { ...VIDEO_USAGE },
@@ -1408,7 +1428,7 @@ var VIDEO_MODELS = [
       logo: GOOGLE_LOGO2,
       description: "Google Gemini Omni Flash 1.1 conversational video generation, editing and extension; text, image, and up to 10-second video input, 3-10 second output at up to 4K and 24 FPS, extendable to 40 seconds."
     },
-    capabilities: { acceptsSourceVideo: true }
+    capabilities: { acceptsSourceVideo: true, lastFrameRequiresFirstFrame: true }
   },
   {
     // vibe-triggers and Space Builder drop the ratio when a frame is wired; Cowork always sends it.
@@ -1434,7 +1454,9 @@ var VIDEO_MODELS = [
       },
       sourceVideoSeconds: { min: 2, max: 15, inReferenceTotal: true },
       ratioIgnoredWith: ["firstFrame", "sourceVideo"],
-      supports: ["generateAudio", "seed", "lastFrame", "watermark"]
+      // No seed: 2.x echoes it but does not honour it (two seeded runs differed, 2026-10-01).
+      supports: ["generateAudio", "lastFrame", "watermark"],
+      audio: "optional"
     },
     defaults: {
       durationSeconds: 5,
@@ -1473,7 +1495,9 @@ var VIDEO_MODELS = [
       },
       sourceVideoSeconds: { min: 2, max: 15, inReferenceTotal: true },
       ratioIgnoredWith: ["firstFrame", "sourceVideo"],
-      supports: ["generateAudio", "seed", "lastFrame", "watermark"]
+      // No seed: 2.x echoes it but does not honour it (two seeded runs differed, 2026-10-01).
+      supports: ["generateAudio", "lastFrame", "watermark"],
+      audio: "optional"
     },
     defaults: {
       durationSeconds: 5,
@@ -1491,7 +1515,6 @@ var VIDEO_MODELS = [
     capabilities: { acceptsSourceVideo: true, ...FRAMES_EXCLUDE_REFERENCES }
   },
   {
-    // generateAudio defaults to false here (trigger-workflows); Space Builder defaults it to true.
     // omni_reference_task_type edit/extend: adaptive ratio, edit sends duration -1.
     id: "seedance-2.5",
     aliases: ["seedance-2-5", "dreamina-seedance-2-5-260628"],
@@ -1513,13 +1536,15 @@ var VIDEO_MODELS = [
       },
       sourceVideoSeconds: { min: 2, max: 30, editMin: 4, inReferenceTotal: true },
       ratioIgnoredWith: ["firstFrame", "sourceVideo"],
-      supports: ["generateAudio", "seed", "lastFrame", "watermark", "taskType"]
+      // No seed: 2.x echoes it but does not honour it (two seeded runs differed, 2026-10-01).
+      supports: ["generateAudio", "lastFrame", "watermark", "taskType"],
+      audio: "optional"
     },
     defaults: {
       durationSeconds: 5,
       resolution: "720p",
       aspectRatio: "16:9",
-      generateAudio: false
+      generateAudio: true
     },
     usage: { ...VIDEO_USAGE },
     pricingKind: "video_per_second",
@@ -1531,7 +1556,6 @@ var VIDEO_MODELS = [
     capabilities: { acceptsSourceVideo: true, ...FRAMES_EXCLUDE_REFERENCES }
   },
   {
-    // Cowork allows 7 reference images and durations 4/6/8/10; Space Builder allows 5 and 1-15s.
     id: "grok-imagine",
     aliases: ["grok-imagine-video", "grokImagine"],
     label: "Grok Imagine",
@@ -1545,8 +1569,11 @@ var VIDEO_MODELS = [
       resolutions: ["480p", "720p"],
       durations: { min: 1, max: 15 },
       durationPresets: [4, 6, 8, 10],
-      references: { images: 5 },
-      supports: []
+      // xAI: "A maximum of 7 reference images can be provided per request."
+      references: { images: 7 },
+      supports: [],
+      // xAI can switch sound off (generate_audio); xai-video.ts never sends it.
+      audio: "always"
     },
     // No resolution default: xAI picks its own unless the caller asks.
     defaults: { durationSeconds: 8, aspectRatio: "16:9" },
@@ -1568,15 +1595,16 @@ var VIDEO_MODELS = [
     providerModelId: "kling-video/v2.5-turbo/pro/image-to-video",
     lifecycle: "active",
     exposure: ["cowork", "space_builder"],
+    // The endpoint has no aspect_ratio or last_image_url: the start image sets the framing.
     constraints: {
-      aspectRatios: [...KLING_RATIOS],
       durations: [5, 10],
       requires: ["firstFrame"],
-      supports: ["cfgScale", "negativePrompt", "lastFrame"]
+      supports: ["cfgScale", "negativePrompt"],
+      prompt: KLING_PROMPT,
+      audio: "never"
     },
     defaults: {
       durationSeconds: 5,
-      aspectRatio: "16:9",
       cfgScale: 0.5,
       negativePrompt: NEGATIVE_PROMPT
     },
@@ -1585,7 +1613,7 @@ var VIDEO_MODELS = [
     display: {
       provider: "kuaishou",
       logo: "https://cdn.jsdelivr.net/npm/@lobehub/icons-static-png@1.97.0/light/kling-color.png",
-      description: "Kuaishou Kling 2.5 Turbo Pro image-to-video; animates a single input image from a text prompt with an optional end image, 5- or 10-second clips."
+      description: "Kuaishou Kling 2.5 Turbo Pro image-to-video; animates a single input image from a text prompt, 5- or 10-second clips."
     },
     capabilities: { ...FRAMES_EXCLUDE_REFERENCES }
   },
@@ -1599,17 +1627,18 @@ var VIDEO_MODELS = [
     providerModelId: "kling-video/v3.0/pro/image-to-video",
     lifecycle: "active",
     exposure: ["cowork", "space_builder"],
+    // The endpoint has no aspect_ratio or negative_prompt; sound is "on"/"off" (default on).
     constraints: {
-      aspectRatios: [...KLING_RATIOS],
       durations: { min: 3, max: 15 },
       requires: ["firstFrame"],
-      supports: ["cfgScale", "negativePrompt", "lastFrame"]
+      supports: ["cfgScale", "lastFrame", "generateAudio"],
+      prompt: KLING_PROMPT,
+      audio: "optional"
     },
     defaults: {
       durationSeconds: 5,
-      aspectRatio: "16:9",
       cfgScale: 0.5,
-      negativePrompt: NEGATIVE_PROMPT
+      generateAudio: true
     },
     usage: { ...VIDEO_USAGE },
     pricingKind: "video_per_second",
@@ -1621,7 +1650,7 @@ var VIDEO_MODELS = [
     capabilities: { ...FRAMES_EXCLUDE_REFERENCES }
   },
   {
-    // mode standard/pro selects 720p/1080p; the endpoint takes no separate resolution.
+    // mode standard/pro/4k selects 720p/1080p/4K; the endpoint takes no separate resolution.
     // Replicate: "Max 7 [images] without video, 4 with video"; 4K does not support
     // reference_video. No `videos` limit yet: the adapter sends no reference video,
     // so `imagesWithVideo` stays dormant until one is wired through.
@@ -1634,21 +1663,25 @@ var VIDEO_MODELS = [
     lifecycle: "active",
     exposure: ["cowork", "space_builder"],
     constraints: {
-      aspectRatios: [...KLING_RATIOS],
-      resolutions: ["720p", "1080p"],
+      aspectRatios: [...KLING_OMNI_RATIOS],
+      resolutions: ["720p", "1080p", "4k"],
       durations: { min: 3, max: 15 },
       references: { images: 7, imagesWithVideo: 4 },
-      supports: ["generateAudio", "lastFrame"]
+      supports: ["generateAudio", "lastFrame"],
+      prompt: KLING_PROMPT,
+      audio: "optional"
     },
-    defaults: { durationSeconds: 5, aspectRatio: "16:9", generateAudio: false, mode: "pro" },
+    defaults: { durationSeconds: 5, aspectRatio: "16:9", generateAudio: true, mode: "pro" },
     usage: { unit: "compute_seconds" },
     pricingKind: "video_per_second",
     display: {
       provider: "kuaishou",
       logo: "https://cdn.jsdelivr.net/npm/@lobehub/icons-static-png@1.97.0/light/kling-color.png",
-      description: "Kuaishou Kling Video 3.0 Omni unified generation and editing from text, start/end images, up to 7 reference images, and reference video; up to 15 seconds at up to 1080p."
-    },
-    capabilities: {}
+      description: "Kuaishou Kling Video 3.0 Omni unified generation and editing from text, start/end images, up to 7 reference images, and reference video; up to 15 seconds at up to 4K."
+    }
+    // end_image "Requires start_image", but no lastFrameRequiresFirstFrame: SpaceLab resolves
+    // replicate rows in error mode, so the flag would refuse its end-frame-only nodes, which
+    // the adapter sends as start_image.
   },
   {
     id: "minimax-h3",
@@ -1663,7 +1696,10 @@ var VIDEO_MODELS = [
       aspectRatios: [...H3_RATIOS],
       resolutions: ["480p", "768p", "2k", "4k"],
       durations: { min: 5, max: 15 },
-      supports: ["seed", "promptExpansion", "lastFrame", "safetyChecker"]
+      // fal's image-to-video endpoint has no aspect_ratio; the frame sets it.
+      ratioIgnoredWith: ["firstFrame"],
+      supports: ["seed", "promptExpansion", "lastFrame", "safetyChecker"],
+      audio: "always"
     },
     defaults: {
       durationSeconds: 5,
@@ -1691,9 +1727,12 @@ var VIDEO_MODELS = [
     exposure: ["cowork", "space_builder"],
     constraints: {
       aspectRatios: [...H3_RATIOS],
-      resolutions: ["480p", "768p"],
+      resolutions: ["480p", "768p", "1080p"],
       durations: { min: 5, max: 15 },
-      supports: ["seed", "promptExpansion", "lastFrame", "safetyChecker"]
+      // fal's image-to-video endpoint has no aspect_ratio; the frame sets it.
+      ratioIgnoredWith: ["firstFrame"],
+      supports: ["seed", "promptExpansion", "lastFrame", "safetyChecker"],
+      audio: "always"
     },
     defaults: {
       durationSeconds: 5,
@@ -1706,7 +1745,7 @@ var VIDEO_MODELS = [
     display: {
       provider: "minimax",
       logo: "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/f4c2e906-28b7-4a4a-b491-df8fef7808ed.png",
-      description: "High-speed MiniMax H3 Max variant for text-to-video and image-to-video (first/last frame) only, up to 768p."
+      description: "High-speed MiniMax H3 Max variant for text-to-video and image-to-video (first/last frame) only, up to 1080p."
     },
     capabilities: { ...FRAMES_EXCLUDE_REFERENCES }
   },
@@ -1723,6 +1762,8 @@ var VIDEO_MODELS = [
       aspectRatios: [...H3_REFERENCE_RATIOS],
       resolutions: ["480p", "768p", "2k", "4k"],
       durations: { min: 5, max: 15 },
+      // A start frame routes to image-to-video, which has no aspect_ratio.
+      ratioIgnoredWith: ["firstFrame"],
       references: {
         images: 9,
         videos: 3,
@@ -1731,7 +1772,8 @@ var VIDEO_MODELS = [
         audioOnly: false,
         videoSeconds: { min: 2, max: 15, total: 15 }
       },
-      supports: ["seed", "promptExpansion", "safetyChecker"]
+      supports: ["seed", "promptExpansion", "safetyChecker"],
+      audio: "always"
     },
     defaults: {
       durationSeconds: 5,
@@ -1746,7 +1788,9 @@ var VIDEO_MODELS = [
       logo: "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/f4c2e906-28b7-4a4a-b491-df8fef7808ed.png",
       description: "MiniMax H3 reference-to-video mode guided by up to 9 reference images, 3 reference videos, and 3 audio clips; reference and image-to-video inputs cannot be mixed."
     },
-    capabilities: {}
+    // fal's reference-to-video endpoint never reads a frame; the resolver must drop the
+    // references instead of the adapter silently dropping the frame.
+    capabilities: { ...FRAMES_EXCLUDE_REFERENCES }
   },
   {
     id: "minimax-h3-max-reference",
@@ -1759,8 +1803,10 @@ var VIDEO_MODELS = [
     exposure: ["cowork", "space_builder"],
     constraints: {
       aspectRatios: [...H3_REFERENCE_RATIOS],
-      resolutions: ["480p", "768p"],
+      resolutions: ["480p", "768p", "1080p"],
       durations: { min: 5, max: 15 },
+      // A start frame routes to image-to-video, which has no aspect_ratio.
+      ratioIgnoredWith: ["firstFrame"],
       references: {
         images: 9,
         videos: 3,
@@ -1769,7 +1815,8 @@ var VIDEO_MODELS = [
         audioOnly: false,
         videoSeconds: { min: 2, max: 15, total: 15 }
       },
-      supports: ["seed", "promptExpansion", "safetyChecker"]
+      supports: ["seed", "promptExpansion", "safetyChecker"],
+      audio: "always"
     },
     defaults: {
       durationSeconds: 5,
@@ -1784,7 +1831,9 @@ var VIDEO_MODELS = [
       logo: "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/f4c2e906-28b7-4a4a-b491-df8fef7808ed.png",
       description: "MiniMax H3 Max reference-to-video on fal; conditions on reference images, videos, and audio. MiniMax's own docs do not list a reference mode for H3 Max."
     },
-    capabilities: {}
+    // fal's reference-to-video endpoint never reads a frame; the resolver must drop the
+    // references instead of the adapter silently dropping the frame.
+    capabilities: { ...FRAMES_EXCLUDE_REFERENCES }
   },
   {
     // Duration 3-10s per trigger-workflows mapDuration; the endpoint takes no resolution.
@@ -1801,7 +1850,8 @@ var VIDEO_MODELS = [
       durations: { min: 3, max: 10 },
       requires: ["sourceVideo"],
       sourceVideoSeconds: { min: 3, max: 10.05 },
-      supports: []
+      supports: [],
+      audio: "never"
     },
     defaults: { durationSeconds: 5, aspectRatio: "16:9" },
     usage: { ...VIDEO_USAGE },
@@ -1825,7 +1875,8 @@ var VIDEO_MODELS = [
     constraints: {
       resolutions: ["768p"],
       durations: [6, 10],
-      supports: ["lastFrame"]
+      supports: ["lastFrame"],
+      audio: "never"
     },
     defaults: { durationSeconds: 6, resolution: "768p" },
     usage: { unit: "compute_seconds" },
@@ -1848,7 +1899,8 @@ var VIDEO_MODELS = [
     exposure: [],
     constraints: {
       requires: ["firstFrame"],
-      supports: ["seed", "lastFrame"]
+      supports: ["seed", "lastFrame"],
+      audio: "never"
     },
     defaults: {},
     usage: { ...VIDEO_USAGE },
@@ -1870,7 +1922,7 @@ var MODELS = [
 ];
 
 // src/catalog/schema.ts
-import { z as z2 } from "zod";
+import { z as z3 } from "zod";
 
 // src/catalog/collisions.ts
 function collisionIssues(rows) {
@@ -1917,8 +1969,61 @@ function collisionIssues(rows) {
   return issues;
 }
 
-// src/catalog/input-spec.ts
+// src/catalog/conventions.ts
 import { z } from "zod";
+var AudioOutputSchema = z.enum(["always", "optional", "never"]);
+var ReferenceSyntaxSchema = z.enum(["numbered", "caption", "descriptive"]);
+var MaxCharsSchema = z.number().int().positive();
+var PromptConventionsSchema = z.strictObject({
+  referenceSyntax: ReferenceSyntaxSchema,
+  maxChars: MaxCharsSchema.optional()
+});
+var AUDIO_OUTPUTS = AudioOutputSchema.options;
+var REFERENCE_SYNTAXES = ReferenceSyntaxSchema.options;
+var ConstraintsReadView = z.looseObject({
+  audio: AudioOutputSchema.optional().catch(void 0),
+  prompt: z.looseObject({
+    referenceSyntax: ReferenceSyntaxSchema.catch("numbered"),
+    maxChars: MaxCharsSchema.optional().catch(void 0)
+  }).optional().catch(void 0)
+});
+var DEFAULT_CONVENTIONS = { referenceSyntax: "numbered" };
+function outputAudio(row) {
+  const view = ConstraintsReadView.safeParse(row.constraints);
+  return view.success ? view.data.audio : void 0;
+}
+function promptConventions(row) {
+  const view = ConstraintsReadView.safeParse(row.constraints);
+  const prompt = view.success ? view.data.prompt : void 0;
+  if (prompt === void 0) return { ...DEFAULT_CONVENTIONS };
+  const result = { referenceSyntax: prompt.referenceSyntax };
+  if (prompt.maxChars !== void 0) result.maxChars = prompt.maxChars;
+  return result;
+}
+var RawConventionsView = z.looseObject({
+  audio: z.unknown().optional(),
+  prompt: z.unknown().optional()
+});
+function normaliseConventions(row) {
+  const view = RawConventionsView.safeParse(row.constraints);
+  if (!view.success) return { row, invalidKeys: [] };
+  const { audio, prompt } = view.data;
+  const badAudio = audio !== void 0 && !AudioOutputSchema.safeParse(audio).success;
+  const badPrompt = prompt !== void 0 && !PromptConventionsSchema.safeParse(prompt).success;
+  if (!badAudio && !badPrompt) return { row, invalidKeys: [] };
+  const invalidKeys = [];
+  if (badAudio) invalidKeys.push("constraints.audio");
+  if (badPrompt) invalidKeys.push("constraints.prompt");
+  const entries = Object.entries(row.constraints).flatMap(([key, value]) => {
+    if (key === "audio" && badAudio) return [];
+    if (key === "prompt" && badPrompt) return [[key, promptConventions(row)]];
+    return [[key, value]];
+  });
+  return { row: { ...row, constraints: Object.fromEntries(entries) }, invalidKeys };
+}
+
+// src/catalog/input-spec.ts
+import { z as z2 } from "zod";
 var VIDEO_INPUT_SPEC_KEYS = [
   "prompt",
   "durationSeconds",
@@ -1957,8 +2062,8 @@ var ALL_KEY_SET = /* @__PURE__ */ new Set([
   ...VIDEO_INPUT_SPEC_KEYS,
   ...AUDIO_INPUT_SPEC_KEYS
 ]);
-var ProviderScalarSchema = z.union([z.string(), z.number(), z.boolean()]);
-var FieldsSchema = z.record(z.string(), z.string().min(1)).superRefine((fields, ctx) => {
+var ProviderScalarSchema = z2.union([z2.string(), z2.number(), z2.boolean()]);
+var FieldsSchema = z2.record(z2.string(), z2.string().min(1)).superRefine((fields, ctx) => {
   const keys = Object.keys(fields);
   if (keys.length === 0) {
     ctx.addIssue({
@@ -1976,7 +2081,7 @@ var FieldsSchema = z.record(z.string(), z.string().min(1)).superRefine((fields,
     }
   }
 });
-var ValuesSchema = z.record(z.string(), z.record(z.string(), ProviderScalarSchema)).superRefine((values, ctx) => {
+var ValuesSchema = z2.record(z2.string(), z2.record(z2.string(), ProviderScalarSchema)).superRefine((values, ctx) => {
   for (const key of Object.keys(values)) {
     if (!ALL_KEY_SET.has(key)) {
       ctx.addIssue({
@@ -1986,8 +2091,8 @@ var ValuesSchema = z.record(z.string(), z.record(z.string(), ProviderScalarSchem
     }
   }
 });
-var StaticSchema = z.record(z.string(), ProviderScalarSchema);
-var InputSpecSchema = z.strictObject({
+var StaticSchema = z2.record(z2.string(), ProviderScalarSchema);
+var InputSpecSchema = z2.strictObject({
   fields: FieldsSchema,
   values: ValuesSchema.optional(),
   static: StaticSchema.optional()
@@ -2091,16 +2196,16 @@ var ADAPTER_IDS = [
   "higgsfield",
   "xai-video"
 ];
-var ModelTypeSchema = z2.enum(["text", "image", "video", "audio", "upscale", "bg-remove"]);
-var LifecycleSchema = z2.enum(["active", "deprecated", "retired"]);
-var SurfaceSchema = z2.enum(["cowork", "space_builder"]);
+var ModelTypeSchema = z3.enum(["text", "image", "video", "audio", "upscale", "bg-remove"]);
+var LifecycleSchema = z3.enum(["active", "deprecated", "retired"]);
+var SurfaceSchema = z3.enum(["cowork", "space_builder"]);
 var SURFACES2 = SurfaceSchema.options;
-var ExposureSchema = z2.preprocess(
+var ExposureSchema = z3.preprocess(
   normalizeExposure,
-  z2.array(SurfaceSchema)
+  z3.array(SurfaceSchema)
 );
-var AdapterSchema = z2.enum(ADAPTER_IDS);
-var AspectRatioSchema = z2.enum([
+var AdapterSchema = z3.enum(ADAPTER_IDS);
+var AspectRatioSchema = z3.enum([
   "1:1",
   "16:9",
   "9:16",
@@ -2115,12 +2220,12 @@ var AspectRatioSchema = z2.enum([
   "97:30",
   "adaptive"
 ]);
-var ImageSizeSchema = z2.enum(["512", "1K", "2K", "4K"]);
-var VideoResolutionSchema = z2.enum(["360p", "480p", "720p", "768p", "1080p", "2k", "4k"]);
-var ImageQualitySchema = z2.enum(["auto", "low", "medium", "high", "xhigh", "max"]);
-var OutputFormatSchema = z2.enum(["png", "jpeg", "webp"]);
-var RequiredInputSchema = z2.enum(["firstFrame", "sourceVideo", "references"]);
-var SupportedOptionSchema = z2.enum([
+var ImageSizeSchema = z3.enum(["512", "1K", "2K", "4K"]);
+var VideoResolutionSchema = z3.enum(["360p", "480p", "720p", "768p", "1080p", "2k", "4k"]);
+var ImageQualitySchema = z3.enum(["auto", "low", "medium", "high", "xhigh", "max"]);
+var OutputFormatSchema = z3.enum(["png", "jpeg", "webp"]);
+var RequiredInputSchema = z3.enum(["firstFrame", "sourceVideo", "references"]);
+var SupportedOptionSchema = z3.enum([
   "seed",
   "quality",
   "negativePrompt",
@@ -2140,142 +2245,147 @@ var SupportedOptionSchema = z2.enum([
   "videoExtension",
   "instrumental"
 ]);
-var UsageUnitSchema = z2.enum([
+var UsageUnitSchema = z3.enum([
   "tokens",
   "images",
   "seconds",
   "reported_cost_usd",
   "compute_seconds"
 ]);
-var PricingKindSchema = z2.enum(["token", "image_size", "video_per_second", "flat"]);
-var ProviderOptionValueSchema = z2.lazy(
-  () => z2.union([
-    z2.string(),
-    z2.number(),
-    z2.boolean(),
-    z2.array(ProviderOptionValueSchema),
-    z2.record(z2.string(), ProviderOptionValueSchema)
+var PricingKindSchema = z3.enum(["token", "image_size", "video_per_second", "flat"]);
+var ProviderOptionValueSchema = z3.lazy(
+  () => z3.union([
+    z3.string(),
+    z3.number(),
+    z3.boolean(),
+    z3.array(ProviderOptionValueSchema),
+    z3.record(z3.string(), ProviderOptionValueSchema)
   ])
 );
-var ProviderOptionsSchema = z2.record(z2.string(), z2.record(z2.string(), ProviderOptionValueSchema));
-var VideoSecondsLimitsSchema = z2.looseObject({
-  min: z2.number().optional(),
-  max: z2.number().optional(),
-  total: z2.number().optional(),
-  editMin: z2.number().optional()
+var ProviderOptionsSchema = z3.record(z3.string(), z3.record(z3.string(), ProviderOptionValueSchema));
+var VideoSecondsLimitsSchema = z3.looseObject({
+  min: z3.number().optional(),
+  max: z3.number().optional(),
+  total: z3.number().optional(),
+  editMin: z3.number().optional()
 });
-var ReferenceLimitsSchema = z2.looseObject({
-  images: z2.number().optional(),
-  imagesWithVideo: z2.number().optional(),
-  videos: z2.number().optional(),
-  audios: z2.number().optional(),
-  total: z2.number().optional(),
-  audioOnly: z2.boolean().optional(),
+var ReferenceLimitsSchema = z3.looseObject({
+  images: z3.number().optional(),
+  imagesWithVideo: z3.number().optional(),
+  videos: z3.number().optional(),
+  audios: z3.number().optional(),
+  total: z3.number().optional(),
+  audioOnly: z3.boolean().optional(),
   videoSeconds: VideoSecondsLimitsSchema.optional()
 });
-var ModelConstraintsSchema = z2.looseObject({
-  aspectRatios: z2.array(AspectRatioSchema).optional(),
-  sizes: z2.array(ImageSizeSchema).optional(),
-  resolutions: z2.array(VideoResolutionSchema).optional(),
-  durations: z2.union([z2.looseObject({ min: z2.number(), max: z2.number() }), z2.array(z2.number())]).optional(),
-  durationPresets: z2.array(z2.number()).optional(),
-  qualityOptions: z2.array(z2.string()).optional(),
-  exactSizes: z2.record(z2.string(), z2.string()).optional(),
+var ModelConstraintsSchema = z3.looseObject({
+  aspectRatios: z3.array(AspectRatioSchema).optional(),
+  sizes: z3.array(ImageSizeSchema).optional(),
+  resolutions: z3.array(VideoResolutionSchema).optional(),
+  durations: z3.union([z3.looseObject({ min: z3.number(), max: z3.number() }), z3.array(z3.number())]).optional(),
+  durationPresets: z3.array(z3.number()).optional(),
+  qualityOptions: z3.array(z3.string()).optional(),
+  exactSizes: z3.record(z3.string(), z3.string()).optional(),
   references: ReferenceLimitsSchema.optional(),
-  requires: z2.array(RequiredInputSchema).optional(),
-  supports: z2.array(SupportedOptionSchema).optional(),
-  ratioIgnoredWith: z2.array(z2.enum(["firstFrame", "sourceVideo"])).optional(),
-  sourceVideoSeconds: VideoSecondsLimitsSchema.omit({ total: true }).extend({ inReferenceTotal: z2.boolean().optional() }).optional(),
-  extension: z2.looseObject({ segmentSeconds: z2.number(), maxTotalSeconds: z2.number() }).optional()
+  requires: z3.array(RequiredInputSchema).optional(),
+  supports: z3.array(SupportedOptionSchema).optional(),
+  ratioIgnoredWith: z3.array(z3.enum(["firstFrame", "sourceVideo"])).optional(),
+  sourceVideoSeconds: VideoSecondsLimitsSchema.omit({ total: true }).extend({ inReferenceTotal: z3.boolean().optional() }).optional(),
+  extension: z3.looseObject({ segmentSeconds: z3.number(), maxTotalSeconds: z3.number() }).optional(),
+  // Loose on purpose: the read path keeps a row whose value is invalid (the accessors in
+  // conventions.ts read it as "not set"); CatalogModelSchema below rejects it on write.
+  audio: z3.unknown().optional(),
+  prompt: z3.unknown().optional()
 });
-var MediaRefSchema = z2.union([
-  z2.looseObject({
-    url: z2.string(),
-    label: z2.string().optional(),
-    durationSeconds: z2.number().optional()
+var MediaRefSchema = z3.union([
+  z3.looseObject({
+    url: z3.string(),
+    label: z3.string().optional(),
+    durationSeconds: z3.number().optional()
   }),
-  z2.looseObject({
-    bytes: z2.instanceof(Uint8Array),
-    mime: z2.string(),
-    label: z2.string().optional(),
-    durationSeconds: z2.number().optional()
+  z3.looseObject({
+    bytes: z3.instanceof(Uint8Array),
+    mime: z3.string(),
+    label: z3.string().optional(),
+    durationSeconds: z3.number().optional()
   })
 ]);
-var ImageSafetySettingSchema = z2.looseObject({ category: z2.string(), threshold: z2.string() });
-var UpscaleOptionsSchema = z2.looseObject({
-  enhanceModel: z2.string().optional(),
-  subjectDetection: z2.string().optional(),
-  faceEnhancement: z2.boolean().optional(),
-  faceEnhancementCreativity: z2.number().optional(),
-  faceEnhancementStrength: z2.number().optional()
+var ImageSafetySettingSchema = z3.looseObject({ category: z3.string(), threshold: z3.string() });
+var UpscaleOptionsSchema = z3.looseObject({
+  enhanceModel: z3.string().optional(),
+  subjectDetection: z3.string().optional(),
+  faceEnhancement: z3.boolean().optional(),
+  faceEnhancementCreativity: z3.number().optional(),
+  faceEnhancementStrength: z3.number().optional()
 });
-var DefaultsSchema = z2.looseObject({
-  references: z2.array(MediaRefSchema).optional(),
+var DefaultsSchema = z3.looseObject({
+  references: z3.array(MediaRefSchema).optional(),
   aspectRatio: AspectRatioSchema.optional(),
   size: ImageSizeSchema.optional(),
-  exactSize: z2.string().optional(),
+  exactSize: z3.string().optional(),
   quality: ImageQualitySchema.optional(),
-  seed: z2.number().optional(),
-  count: z2.number().optional(),
+  seed: z3.number().optional(),
+  count: z3.number().optional(),
   outputFormat: OutputFormatSchema.optional(),
-  scaleFactor: z2.number().optional(),
-  safetySettings: z2.array(ImageSafetySettingSchema).optional(),
-  temperature: z2.number().optional(),
+  scaleFactor: z3.number().optional(),
+  safetySettings: z3.array(ImageSafetySettingSchema).optional(),
+  temperature: z3.number().optional(),
   upscale: UpscaleOptionsSchema.optional(),
   firstFrame: MediaRefSchema.optional(),
   lastFrame: MediaRefSchema.optional(),
   sourceVideo: MediaRefSchema.optional(),
-  referenceImages: z2.array(MediaRefSchema).optional(),
-  referenceVideos: z2.array(MediaRefSchema).optional(),
-  referenceAudios: z2.array(MediaRefSchema).optional(),
+  referenceImages: z3.array(MediaRefSchema).optional(),
+  referenceVideos: z3.array(MediaRefSchema).optional(),
+  referenceAudios: z3.array(MediaRefSchema).optional(),
   resolution: VideoResolutionSchema.optional(),
-  durationSeconds: z2.number().optional(),
-  generateAudio: z2.boolean().optional(),
-  negativePrompt: z2.string().optional(),
-  cfgScale: z2.number().optional(),
-  promptExpansion: z2.enum(["fast", "balanced", "quality"]).optional(),
-  watermark: z2.boolean().optional(),
-  safetyChecker: z2.boolean().optional(),
-  taskType: z2.enum(["reference", "edit", "extend"]).optional(),
-  mode: z2.enum(["standard", "pro"]).optional(),
-  instrumental: z2.boolean().optional()
+  durationSeconds: z3.number().optional(),
+  generateAudio: z3.boolean().optional(),
+  negativePrompt: z3.string().optional(),
+  cfgScale: z3.number().optional(),
+  promptExpansion: z3.enum(["fast", "balanced", "quality"]).optional(),
+  watermark: z3.boolean().optional(),
+  safetyChecker: z3.boolean().optional(),
+  taskType: z3.enum(["reference", "edit", "extend"]).optional(),
+  mode: z3.enum(["standard", "pro"]).optional(),
+  instrumental: z3.boolean().optional()
 });
-var CatalogDisplaySchema = z2.looseObject({
-  provider: z2.string(),
-  description: z2.string(),
-  logo: z2.string().optional(),
-  costTier: z2.union([z2.literal(1), z2.literal(2), z2.literal(3)]).optional(),
-  recommended: z2.boolean().optional(),
-  sortOrder: z2.number().optional(),
-  chat: z2.looseObject({
+var CatalogDisplaySchema = z3.looseObject({
+  provider: z3.string(),
+  description: z3.string(),
+  logo: z3.string().optional(),
+  costTier: z3.union([z3.literal(1), z3.literal(2), z3.literal(3)]).optional(),
+  recommended: z3.boolean().optional(),
+  sortOrder: z3.number().optional(),
+  chat: z3.looseObject({
     providerOptions: ProviderOptionsSchema.optional(),
-    temperature: z2.number().optional()
+    temperature: z3.number().optional()
   }).optional()
 });
-var CatalogCapabilitiesSchema = z2.looseObject({
-  acceptsSourceVideo: z2.boolean().optional(),
-  framesExcludeReferences: z2.boolean().optional(),
-  reasoning: z2.boolean().optional(),
-  vision: z2.boolean().optional(),
-  contextWindow: z2.number().optional()
+var CatalogCapabilitiesSchema = z3.looseObject({
+  acceptsSourceVideo: z3.boolean().optional(),
+  framesExcludeReferences: z3.boolean().optional(),
+  lastFrameRequiresFirstFrame: z3.boolean().optional(),
+  reasoning: z3.boolean().optional(),
+  vision: z3.boolean().optional(),
+  contextWindow: z3.number().optional()
 });
-var CatalogModelSchemaBase = z2.looseObject({
-  id: z2.string().min(1),
-  aliases: z2.array(z2.string()).default([]),
-  label: z2.string().min(1),
+var CatalogModelSchemaBase = z3.looseObject({
+  id: z3.string().min(1),
+  aliases: z3.array(z3.string()).default([]),
+  label: z3.string().min(1),
   type: ModelTypeSchema,
   adapter: AdapterSchema,
-  providerModelId: z2.string().min(1),
+  providerModelId: z3.string().min(1),
   lifecycle: LifecycleSchema,
-  replacedBy: z2.string().optional(),
+  replacedBy: z3.string().optional(),
   exposure: ExposureSchema,
   constraints: ModelConstraintsSchema,
   defaults: DefaultsSchema,
   providerOptions: ProviderOptionsSchema.optional(),
-  usage: z2.looseObject({
+  usage: z3.looseObject({
     unit: UsageUnitSchema,
-    slug: z2.string().optional(),
-    providerLabel: z2.string().optional()
+    slug: z3.string().optional(),
+    providerLabel: z3.string().optional()
   }),
   pricingKind: PricingKindSchema,
   display: CatalogDisplaySchema.optional(),
@@ -2326,9 +2436,30 @@ var CatalogModelSchema = CatalogModelSchemaBase.superRefine(
         }
       }
     }
+    const audio = row.constraints.audio;
+    if (audio !== void 0 && !AudioOutputSchema.safeParse(audio).success) {
+      ctx.addIssue({
+        code: "custom",
+        path: ["constraints", "audio"],
+        message: 'audio must be "always", "optional" or "never"'
+      });
+    }
+    const prompt = row.constraints.prompt;
+    if (prompt !== void 0) {
+      const parsed = PromptConventionsSchema.safeParse(prompt);
+      if (!parsed.success) {
+        for (const issue of parsed.error.issues) {
+          ctx.addIssue({
+            code: "custom",
+            path: ["constraints", "prompt", ...issue.path.map(String)],
+            message: issue.message
+          });
+        }
+      }
+    }
   }
 );
-var CatalogSchema = z2.array(CatalogModelSchemaBase).superRefine((rows, ctx) => {
+var CatalogSchema = z3.array(CatalogModelSchemaBase.transform((row) => normaliseConventions(row).row)).superRefine((rows, ctx) => {
   for (const issue of collisionIssues(rows)) {
     ctx.addIssue({
       code: "custom",
@@ -2351,42 +2482,42 @@ var liveState = store[KEY] ??= {
 var liveCatalog = liveState.catalog;
 
 // src/json.ts
-import { z as z3 } from "zod";
-var JsonValueSchema = z3.lazy(
-  () => z3.union([
-    z3.string(),
-    z3.number(),
-    z3.boolean(),
-    z3.null(),
-    z3.array(JsonValueSchema),
-    z3.record(z3.string(), JsonValueSchema)
+import { z as z4 } from "zod";
+var JsonValueSchema = z4.lazy(
+  () => z4.union([
+    z4.string(),
+    z4.number(),
+    z4.boolean(),
+    z4.null(),
+    z4.array(JsonValueSchema),
+    z4.record(z4.string(), JsonValueSchema)
   ])
 );
-var JsonObjectSchema = z3.record(z3.string(), JsonValueSchema);
-var thrownErrorSchema = z3.object({
-  message: z3.string().optional(),
-  status: z3.number().optional(),
-  statusCode: z3.number().optional(),
-  responseBody: z3.string().optional()
+var JsonObjectSchema = z4.record(z4.string(), JsonValueSchema);
+var thrownErrorSchema = z4.object({
+  message: z4.string().optional(),
+  status: z4.number().optional(),
+  statusCode: z4.number().optional(),
+  responseBody: z4.string().optional()
 });
 
 // src/catalog/load.ts
-import { z as z4 } from "zod";
+import { z as z5 } from "zod";
 function rowCount() {
   return liveCatalog.list({ lifecycle: ["active", "deprecated", "retired"] }).length;
 }
 function authHeaders(auth) {
   return "bearer" in auth ? { Authorization: `Bearer ${auth.bearer}` } : { "x-api-key": auth.apiKey };
 }
-var EnvelopeSchema = z4.object({
-  rows: z4.array(JsonValueSchema),
-  revision: z4.string().optional()
+var EnvelopeSchema = z5.object({
+  rows: z5.array(JsonValueSchema),
+  revision: z5.string().optional()
 });
 function dropReason(path, message) {
   return path.length === 0 ? message : `${path.map((part) => String(part)).join(".")} ${message}`;
 }
 function rowId(raw) {
-  const parsed = z4.object({ id: z4.string().min(1) }).loose().safeParse(raw);
+  const parsed = z5.object({ id: z5.string().min(1) }).loose().safeParse(raw);
   if (!parsed.success) return "?";
   return parsed.data.id;
 }
@@ -2414,7 +2545,13 @@ function quarantine(rawRows, logger) {
         );
       }
     }
-    valid.push(row);
+    const normalised = normaliseConventions(row);
+    if (normalised.invalidKeys.length > 0) {
+      logger.warn(
+        `[modelkit] catalog row ${row.id} has an invalid ${normalised.invalidKeys.join(", ")}; read as not set`
+      );
+    }
+    valid.push(normalised.row);
   }
   let kept = valid;
   for (; ; ) {
@@ -2809,6 +2946,24 @@ function resolveVideoOptions(model, request, onUnsupported = "error") {
     });
     delete merged.sourceVideo;
   }
+  if (merged.lastFrame !== void 0 && merged.firstFrame === void 0 && model.capabilities?.lastFrameRequiresFirstFrame) {
+    if (onUnsupported === "error") {
+      throw new ModelKitError({
+        code: "missing_input",
+        message: `${model.id} takes an end frame only together with a start frame`,
+        model: model.id,
+        retryable: false
+      });
+    }
+    warnings.push({
+      code: "option_ignored",
+      field: "lastFrame",
+      to: "firstFrame",
+      message: `end frame used as start frame: ${model.id} needs a start frame with an end frame`
+    });
+    merged.firstFrame = merged.lastFrame;
+    delete merged.lastFrame;
+  }
   const hasFrames = merged.firstFrame !== void 0 || merged.lastFrame !== void 0;
   const hasRefs = compact(merged.referenceImages).length + compact(merged.referenceVideos).length + compact(merged.referenceAudios).length > 0;
   if (hasFrames && hasRefs && model.capabilities?.framesExcludeReferences) {
@@ -3028,6 +3183,7 @@ function resolveOptions(model, request, onUnsupported = "error") {
 }
 export {
   ADAPTER_IDS,
+  AUDIO_OUTPUTS,
   Catalog,
   CatalogModelSchema,
   CatalogModelSchemaBase,
@@ -3037,6 +3193,7 @@ export {
   MODELS,
   ModelKitError,
   OTHER_MODELS,
+  REFERENCE_SYNTAXES,
   SURFACES2 as SURFACES,
   SurfaceSchema,
   TEXT_MODELS,
@@ -3049,7 +3206,9 @@ export {
   liveState,
   nearestAspectRatio,
   normalizeExposure,
+  outputAudio,
   primeCatalog,
+  promptConventions,
   resetLiveCatalog,
   resolveAudioOptions,
   resolveImageOptions,