ottoport 1.7.1 → 1.8.0

package/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "ottoport",
   "description": "One API for every model. Call chat, image, video, speech and music models through the OttoPort gateway — as MCP tools, slash commands, or the bundled CLI.",
-  "version": "1.7.1",
+  "version": "1.8.0",
   "author": {
     "name": "LITBOX LLC",
     "email": "support@ottoport.ai"

package/cli/ottoport.mjs

@@ -5,6 +5,7 @@
 //
 //   ottoport models [--modality chat|image|video]
 //   ottoport chat   "prompt" [--model claude-sonnet-5] [--system "..."] [--no-stream]
+//                   [--video <url|file>]   ask about a video (defaults to video-understanding)
 //   ottoport image  "prompt" [--model gpt-image-2] [--resolution 2k] [--ref a.png,b.png] [--out img.png]
 //   ottoport video  "prompt" [--model kling-3.0] [--duration 5] [--out clip.mp4]
 //   ottoport speech "text" [--model gpt-4o-mini-tts] [--voice alloy] [--out speech.mp3]
@@ -17,7 +18,7 @@
 //   The public CLI always targets https://ottoport.ai unless --url is passed
 //   explicitly for local development or a self-hosted gateway.
 
-import { writeFile } from "node:fs/promises";
+import { readFile, stat, writeFile } from "node:fs/promises";
 
 const BASE = "https://ottoport.ai";
 
@@ -152,11 +153,13 @@ async function cmdChat(prompt, flags) {
   const { baseUrl, apiKey } = config(flags);
   const messages = [];
   if (flags.system) messages.push({ role: "system", content: flags.system });
-  messages.push({ role: "user", content: prompt });
+  const videos = await videoParts(flags);
+  messages.push({ role: "user", content: videos.length ? [{ type: "text", text: prompt }, ...videos] : prompt });
 
   const stream = flags.stream !== false;
   const body = {
-    model: last(flags.model) || "claude-sonnet-5",
+    // A video question goes to the model built for it unless one is named.
+    model: last(flags.model) || (videos.length ? "video-understanding" : "claude-sonnet-5"),
     messages,
     stream,
     ...(flags.temperature ? { temperature: Number(last(flags.temperature)) } : {}),
@@ -202,6 +205,39 @@ async function cmdChat(prompt, flags) {
   process.stdout.write("\n");
 }
 
+const VIDEO_TYPES = { ".mp4": "video/mp4", ".m4v": "video/mp4", ".mov": "video/quicktime", ".webm": "video/webm", ".mpeg": "video/mpeg", ".mpg": "video/mpeg" };
+
+/**
+ * `--video`, once or several times (or comma-separated), as chat content parts.
+ *
+ * A URL is sent as-is: the gateway's transport fetches it, so it must be
+ * publicly reachable. A local file is inlined as a base64 data URL, which is
+ * convenient for trying a clip but costs a third more than the file on the
+ * wire — for anything large, host it and pass the URL.
+ */
+async function videoParts(flags) {
+  const list = (Array.isArray(flags.video) ? flags.video : [flags.video])
+    .filter((value) => typeof value === "string")
+    .flatMap((value) => value.split(",").map((one) => one.trim()))
+    .filter(Boolean);
+  const parts = [];
+  for (const source of list) {
+    if (/^(https?:|data:)/i.test(source)) {
+      parts.push({ type: "video_url", video_url: { url: source } });
+      continue;
+    }
+    const info = await stat(source).catch(() => null);
+    if (!info?.isFile()) die(`--video ${source}: not a URL and no such file`);
+    const ext = source.slice(source.lastIndexOf(".")).toLowerCase();
+    const type = VIDEO_TYPES[ext];
+    if (!type) die(`--video ${source}: unsupported type ${ext || "(none)"}; use ${Object.keys(VIDEO_TYPES).join(", ")}`);
+    const mb = info.size / 1e6;
+    if (mb > 20) console.error(`ottoport: ${source} is ${mb.toFixed(0)}MB; inlined it will be ~${(mb * 4 / 3).toFixed(0)}MB and may be refused — host it and pass the URL instead`);
+    parts.push({ type: "video_url", video_url: { url: `data:${type};base64,${(await readFile(source)).toString("base64")}` } });
+  }
+  return parts;
+}
+
 async function download(url, out) {
   const res = await fetch(url);
   if (!res.ok) die(`could not download ${url}: ${res.status}`);
@@ -377,6 +413,7 @@ Usage:
   ottoport models [--modality chat|image|video|tts|music] [--json]
   ottoport models <model-id> [--json]     what that one model accepts
   ottoport chat  "<prompt>" [--model claude-sonnet-5] [--system "..."] [--no-stream]
+                            [--video <url|file>[,<url|file>…]]   ask about a video
                             [--temperature 0.7] [--max-tokens 512]
   ottoport image "<prompt>" [--model gpt-image-2] [--size 1024x1024 | --resolution 2k] [--quality high]
                             [--aspect-ratio 16:9] [--n 1] [--seed 7]
@@ -409,6 +446,7 @@ Examples:
   ottoport models --modality image      # prints what each one accepts
   ottoport models veo-3.1               # its modes, tiers, and every parameter
   ottoport chat "explain MCP in one line" --model claude-haiku-4.5
+  ottoport chat "break this clip down shot by shot" --video https://example.com/clip.mp4
   ottoport image "isometric city at dusk" --model gpt-image-2 --out city.png
   ottoport video "cinematic product reveal" --model seedance-2.0-fast --resolution 1080p
   ottoport video "the logo unfolds" --image start.png --last-frame end.png

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ottoport",
-  "version": "1.7.1",
+  "version": "1.8.0",
   "description": "Claude Code plugin, CLI and MCP server for OttoPort — one OpenAI-compatible API for every LLM, image, video, and speech model.",
   "homepage": "https://ottoport.ai",
   "repository": {