@viralspin/mcp 0.1.0 → 0.2.0

package/src/tools/video-assembly.ts

@@ -0,0 +1,205 @@
+import { spawn } from "child_process";
+import path from "path";
+import { get, post } from "../client.js";
+
+// The assembly engine ships inside this package (mcp/engine). Override with env if needed.
+// __dirname at runtime = mcp/dist/tools → ../../engine = mcp/engine.
+const ENGINE_DIR =
+  process.env.VIRALSPIN_ASSEMBLY_DIR || path.resolve(__dirname, "../../engine");
+
+const BROLL_STYLE_KEYS = ["sketchbook", "cinematic_minimal", "dark_studio", "vibrant_illustrated"] as const;
+const CAPTION_STYLE_KEYS = ["auto", "default", "hormozi", "glow_neon", "minimalist", "whiteboard"] as const;
+
+interface AssemblyArgs {
+  aroll: string;
+  broll?: string[];
+  prompt?: string;
+  output_format?: "9x16" | "1x1" | "16x9";
+  out?: string;
+  google_picker_session?: string;
+  style_ref?: string;
+  broll_mode?: "user_provided" | "ai_generated";
+  style?: typeof BROLL_STYLE_KEYS[number];
+  explainer_index_id?: string;
+  caption_style?: typeof CAPTION_STYLE_KEYS[number];
+}
+
+/** Run the engine orchestrator, streaming its logs to stderr, and capture RESULT:<path>. */
+function runEngine(args: string[], extraEnv: Record<string, string> = {}): Promise<string> {
+  return new Promise((resolve, reject) => {
+    const child = spawn(process.platform === "win32" ? "node.exe" : "node",
+      [path.join(ENGINE_DIR, "assemble.mjs"), ...args],
+      { cwd: ENGINE_DIR, env: { ...process.env, ...extraEnv } });
+    let stdout = "";
+    child.stdout.on("data", (d) => { stdout += d.toString(); });
+    child.stderr.on("data", (d) => process.stderr.write(d)); // progress → MCP server log
+    child.on("error", reject);
+    child.on("close", (code) => {
+      const m = stdout.match(/RESULT:(.+)/);
+      if (code === 0 && m) resolve(m[1].trim());
+      else reject(new Error(`Assembly engine exited ${code}. ${stdout.slice(-400)}`));
+    });
+  });
+}
+
+export const createVideoAssemblyTool = {
+  name: "create_video_assembly",
+  description:
+    "Turn an A-roll talking-head/performance clip into a polished vertical (9:16) short by editing in B-roll cutaways. " +
+    "Runs a LOCAL pipeline: transcribes the A-roll, finds each B-roll's focal point (face/subject-aware so nothing is cropped off), " +
+    "matches B-roll to what's being said, and renders with Ken Burns motion + the speaker's voice as the soundtrack (B-roll clip audio is mixed in as a low faded bed). " +
+    "Accepts LOCAL file paths OR public URLs for both A-roll and B-roll, and a B-roll FOLDER is expanded automatically. " +
+    "Pass style_ref (a reference video) to AUTO-INFER the editing style from it — captions, hook headline, keyword pops, pace, and aspect ratio — so you don't have to describe them. " +
+    "Returns the path to the finished MP4, which you can then publish with publish_now / schedule_post.",
+  inputSchema: {
+    type: "object",
+    properties: {
+      aroll: {
+        type: "string",
+        description: "A-roll source: a local file path (e.g. C:/clips/talk.mp4) OR a public video URL.",
+      },
+      broll: {
+        type: "array",
+        items: { type: "string" },
+        description:
+          "Optional B-roll sources: local file paths, a local FOLDER (expanded to its images/videos), and/or public URLs. " +
+          "Mix freely. Omit for an A-roll-only edit (just motion, no cutaways).",
+      },
+      prompt: {
+        type: "string",
+        description: "Optional creative direction for the edit (mood, what to emphasise, pacing).",
+      },
+      output_format: {
+        type: "string",
+        enum: ["9x16", "1x1", "16x9"],
+        description: "Aspect ratio. Default 9x16 (vertical for Reels/TikTok/Shorts).",
+      },
+      out: {
+        type: "string",
+        description: "Optional output MP4 path. Defaults to the engine's out/ folder.",
+      },
+      google_picker_session: {
+        type: "string",
+        description:
+          "Optional sessionId from start_photo_picker. If set, the photos/clips you picked in Google Photos are ingested as B-roll (in addition to any `broll`).",
+      },
+      style_ref: {
+        type: "string",
+        description:
+          "Optional reference video (local path or URL). The tool watches it and infers the edit style — captions on/off + style, hook headline, keyword pops, cut pace, and aspect ratio — and applies it automatically. Overrides output_format when set.",
+      },
+      broll_mode: {
+        type: "string",
+        enum: ["user_provided", "ai_generated"],
+        description:
+          "B-roll source mode. 'user_provided' (default) = use local files or Google Photos. 'ai_generated' = AI reads the transcript and generates stylized cinematic B-roll images — no local footage needed. Requires the 'style' param.",
+      },
+      style: {
+        type: "string",
+        enum: BROLL_STYLE_KEYS,
+        description:
+          "Style anchor for AI-generated B-roll (required when broll_mode='ai_generated'). " +
+          "sketchbook = hand-drawn ink on graph paper; " +
+          "cinematic_minimal = flat-lay dark surface, warm tungsten lamp; " +
+          "dark_studio = matte black workspace, designer aesthetic; " +
+          "vibrant_illustrated = bold pop-art flat illustration.",
+      },
+      explainer_index_id: {
+        type: "string",
+        description:
+          "Optional Twelve Labs index ID returned by index_explainer_sources. " +
+          "When set, long-form videos (>2 min) in the broll folder are searched semantically " +
+          "instead of being clipped from their start. Requires TWELVE_LABS_API_KEY in the MCP env.",
+      },
+      caption_style: {
+        type: "string",
+        enum: CAPTION_STYLE_KEYS,
+        description:
+          "Caption style. Default 'auto' = AI picks the best style from the transcript. " +
+          "default = bold white spring pop-in. " +
+          "hormozi = word-by-word yellow highlight (high-energy talking-head). " +
+          "glow_neon = neon cyan pulsing text (dark aesthetic, music, gaming). " +
+          "minimalist = lowercase thin font fade-in (calm, educational). " +
+          "whiteboard = left-to-right text reveal (explainers, tutorials). " +
+          "After rendering, change it instantly with swap_caption_style.",
+      },
+    },
+    required: ["aroll"],
+  },
+
+  async execute(args: AssemblyArgs): Promise<string> {
+    // AI-generated B-roll mode: delegate to the cloud narrative pipeline.
+    if (args.broll_mode === "ai_generated") {
+      if (!args.style) {
+        return "Error: 'style' is required when broll_mode='ai_generated'. Choose one of: sketchbook, cinematic_minimal, dark_studio, vibrant_illustrated.";
+      }
+      const job = await post<{ job_id: string; status: string }>("/v1/narrative/jobs", {
+        aroll_paths: [args.aroll],
+        prompt: args.prompt,
+        output_format: args.output_format ?? "9x16",
+        broll_mode: "ai_generated",
+        style_anchor: args.style,
+      });
+      const deadline = Date.now() + 15 * 60 * 1000;
+      let result: { status: string; preview_url?: string | null; download_url?: string | null; error_message?: string | null } = { status: "queued" };
+      while (Date.now() < deadline) {
+        await new Promise((r) => setTimeout(r, 6000));
+        result = await get(`/v1/narrative/jobs/${job.job_id}`);
+        if (result.status === "complete" || result.status === "failed") break;
+      }
+      if (result.status === "failed") return `Assembly failed: ${result.error_message ?? "unknown error"}`;
+      if (result.status !== "complete") return `Assembly timed out. Job ID: ${job.job_id} — check the dashboard.`;
+      return (
+        `AI B-roll assembly complete (style: ${args.style}).\n` +
+        `Preview: ${result.preview_url ?? "—"}\n` +
+        `Download: ${result.download_url ?? "—"}\n` +
+        `Publish it with publish_now or schedule_post.`
+      );
+    }
+
+    const broll = [...(args.broll ?? [])];
+
+    // Pull in Google Photos picks (re-hosted to public URLs by the backend), if any.
+    if (args.google_picker_session) {
+      const r = await get<{ status: string; items?: { url: string }[] }>(
+        `/v1/assembly/picker?sessionId=${encodeURIComponent(args.google_picker_session)}`
+      );
+      if (r.status !== "ready") {
+        return "Your Google Photos picks aren't ready yet — finish selecting in the picker tab, then run create_video_assembly again with the same google_picker_session.";
+      }
+      for (const it of r.items ?? []) broll.push(it.url);
+    }
+
+    const cli: string[] = ["--aroll", args.aroll];
+    if (broll.length) cli.push("--broll", ...broll);
+    if (args.prompt) cli.push("--prompt", args.prompt);
+    if (args.style_ref) cli.push("--style-ref", args.style_ref);
+    if (args.output_format) cli.push("--format", args.output_format);
+    if (args.out) cli.push("--out", args.out);
+    if (args.caption_style && args.caption_style !== "auto") cli.push("--caption-style", args.caption_style);
+
+    // Pull the admin-managed "secret sauce" prompts (LLM Prompt Store →
+    // LOCAL VIDEO ASSEMBLY section) and hand them to the engine. Best-effort:
+    // on any failure the engine falls back to its built-in prompts.
+    const extraEnv: Record<string, string> = {};
+    try {
+      const r = await get<{ prompts?: Record<string, string> }>("/v1/assembly/prompts");
+      if (r.prompts && Object.keys(r.prompts).length) {
+        extraEnv.VIRALSPIN_ASSEMBLY_PROMPTS = JSON.stringify(r.prompts);
+      }
+    } catch { /* engine uses built-in prompts */ }
+    if (args.explainer_index_id) {
+      extraEnv.EXPLAINER_INDEX_ID = args.explainer_index_id;
+    }
+    if (process.env.TWELVE_LABS_API_KEY) {
+      extraEnv.TWELVE_LABS_API_KEY = process.env.TWELVE_LABS_API_KEY;
+    }
+
+    const outPath = await runEngine(cli, extraEnv);
+    return (
+      `Video assembly complete.\n` +
+      `Output: ${outPath}\n` +
+      `Publish it with publish_now (paths: ["${outPath}"]) or schedule_post.`
+    );
+  },
+};

package/tsconfig.json

@@ -1,15 +1,15 @@
-{
-  "compilerOptions": {
-    "target": "ES2020",
-    "module": "commonjs",
-    "lib": ["ES2020"],
-    "outDir": "./dist",
-    "rootDir": "./src",
-    "strict": true,
-    "esModuleInterop": true,
-    "skipLibCheck": true,
-    "resolveJsonModule": true
-  },
-  "include": ["src/**/*"],
-  "exclude": ["node_modules", "dist"]
-}
+{
+  "compilerOptions": {
+    "target": "ES2020",
+    "module": "commonjs",
+    "lib": ["ES2020"],
+    "outDir": "./dist",
+    "rootDir": "./src",
+    "strict": true,
+    "esModuleInterop": true,
+    "skipLibCheck": true,
+    "resolveJsonModule": true
+  },
+  "include": ["src/**/*"],
+  "exclude": ["node_modules", "dist"]
+}