sogni-gen 1.5.2 → 1.5.3

package/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: sogni-gen
-version: "1.5.2"
+version: "1.5.3"
 description: Generate images **and videos** using Sogni AI's decentralized network. Ask the agent to "draw", "generate", "create an image", or "make a video/animate" from a prompt or reference image.
 homepage: https://sogni.ai
 metadata:
@@ -128,6 +128,9 @@ node sogni-gen.mjs -q -o /tmp/cat.png "a cat wearing a hat"
 | `--json` | JSON output | false |
 | `--strict-size` | Do not auto-adjust i2v video size for reference resizing constraints | false |
 | `-q, --quiet` | No progress output | false |
+| `--extract-last-frame <video> <image>` | Extract last frame from video (safe ffmpeg wrapper) | - |
+| `--concat-videos <out> <clips...>` | Concatenate video clips (safe ffmpeg wrapper) | - |
+| `--list-media [type]` | List recent inbound media (images\|audio\|all) | images |
 
 ## OpenClaw Config Defaults
 
@@ -309,7 +312,7 @@ When a user requests a "360 video", follow this workflow:
 
 ### Transition Video Rule
 
-For **any transition video work**, always use the **Sogni skill/plugin** (not ffmpeg or other methods) unless explicitly told otherwise.
+For **any transition video work**, always use the **Sogni skill/plugin** (not raw ffmpeg or other shell commands). Use the built-in `--extract-last-frame`, `--concat-videos`, and `--looping` flags for video manipulation.
 
 ### Insufficient Funds Handling
 
@@ -385,10 +388,11 @@ sogni-gen -c old_photo.jpg -o restored.png -w 1024 -h 1280 \
 
 **Finding received images (Telegram/etc):**
 ```bash
-ls -la ~/.clawdbot/media/inbound/*.jpg | tail -3
-cp ~/.clawdbot/media/inbound/<latest>.jpg /tmp/to_restore.jpg
+node {{skillDir}}/sogni-gen.mjs --json --list-media images
 ```
 
+**Do NOT use `ls`, `cp`, or other shell commands to browse user files.** Always use `--list-media` to find inbound media.
+
 ## IMPORTANT KEYWORD RULE
 
 - If the user message includes the word "photobooth" (case-insensitive), always use `--photobooth` mode with `--ref` set to the user-provided face image.
@@ -417,9 +421,14 @@ node {{skillDir}}/sogni-gen.mjs -q --photobooth --ref /path/to/face.jpg -o /tmp/
 # Check current SPARK/SOGNI balances (no prompt required)
 node {{skillDir}}/sogni-gen.mjs --json --balance
 
+# Find user-sent images/audio
+node {{skillDir}}/sogni-gen.mjs --json --list-media images
+
 # Then send via message tool with filePath
 ```
 
+**Security:** Agents must use the CLI's built-in flags (`--extract-last-frame`, `--concat-videos`, `--list-media`) for all file operations and video manipulation. Never run raw shell commands (`ffmpeg`, `ls`, `cp`, etc.) directly.
+
 ## Animate Between Two Images (First-Frame / Last-Frame)
 
 When a user asks to **animate between two images**, use `--ref` (first frame) and `--ref-end` (last frame) to create a creative interpolation video:
@@ -433,23 +442,23 @@ node {{skillDir}}/sogni-gen.mjs -q --video --ref /tmp/imageA.png --ref-end /tmp/
 
 When a user asks to **animate from a video to an image** (or "continue" a video into a new scene):
 
-1. **Extract the last frame** of the existing video:
+1. **Extract the last frame** of the existing video using the built-in safe wrapper:
    ```bash
-   ffmpeg -y -sseof -0.1 -i /tmp/existing.mp4 -frames:v 1 -update 1 /tmp/lastframe.png
+   node {{skillDir}}/sogni-gen.mjs --extract-last-frame /tmp/existing.mp4 /tmp/lastframe.png
    ```
 2. **Generate a new video** using the last frame as `--ref` and the target image as `--ref-end`:
    ```bash
    node {{skillDir}}/sogni-gen.mjs -q --video --ref /tmp/lastframe.png --ref-end /tmp/target.png -o /tmp/continuation.mp4 "scene transition prompt"
    ```
-3. **Stitch the videos** together with ffmpeg:
+3. **Concatenate the videos** using the built-in safe wrapper:
    ```bash
-   ffmpeg -y -i /tmp/existing.mp4 -i /tmp/continuation.mp4 \
-     -filter_complex "[0:v][1:v]concat=n=2:v=1:a=0[outv]" \
-     -map "[outv]" -c:v libx264 -crf 18 /tmp/full_sequence.mp4
+   node {{skillDir}}/sogni-gen.mjs --concat-videos /tmp/full_sequence.mp4 /tmp/existing.mp4 /tmp/continuation.mp4
    ```
 
 This ensures visual continuity — the new clip picks up exactly where the previous one ended.
 
+**Do NOT run raw `ffmpeg` commands.** Always use `--extract-last-frame` and `--concat-videos` for video manipulation.
+
 **Always apply this pattern when:**
 - User says "animate image A to image B" → use `--ref A --ref-end B`
 - User says "animate this video to this image" → extract last frame, use as `--ref`, target image as `--ref-end`, then stitch

package/llm.txt

@@ -135,6 +135,9 @@ node {{skillDir}}/sogni-gen.mjs --json --balance
 | --last-image | Reuse last generated image as input |
 | --json | Machine-readable JSON output |
 | --balance | Show Spark/Sogni token balances |
+| --extract-last-frame VIDEO IMAGE | Extract last frame from a video file |
+| --concat-videos OUTPUT CLIPS... | Concatenate multiple video clips |
+| --list-media [images\|audio\|all] | List recent inbound media files |
 
 ## Agent Behavior Guidelines
 
@@ -147,22 +150,26 @@ node {{skillDir}}/sogni-gen.mjs --json --balance
 6. Always use -q (quiet) and -o (output to file) so you can send the result back.
 7. After generating, send the file to the user via the message tool with filePath.
 8. If you get "Insufficient funds", tell them: "Claim 50 free daily Spark points at https://app.sogni.ai/"
-9. For transition/animation videos, always use this plugin (not ffmpeg) unless told otherwise.
+9. For transition/animation videos, always use this plugin's built-in flags (not raw ffmpeg). Use `--looping`, `--extract-last-frame`, or `--concat-videos`.
 10. Default to 768x768 for images. Video sizes must be divisible by 16 (min 480px, max 1536px).
 
 ## Finding User-Sent Media
 
-When users send images/audio via Telegram, WhatsApp, or iMessages:
+When users send images/audio via Telegram, WhatsApp, or iMessages, use the built-in `--list-media` flag:
 
 ```bash
-# Recent inbound images
-ls -la ~/.clawdbot/media/inbound/*.jpg | tail -3
-ls -la ~/.clawdbot/media/inbound/*.png | tail -3
+# Recent inbound images (default)
+node {{skillDir}}/sogni-gen.mjs --json --list-media images
 
 # Recent inbound audio
-ls -la ~/.clawdbot/media/inbound/*.m4a | tail -3
+node {{skillDir}}/sogni-gen.mjs --json --list-media audio
+
+# All recent media
+node {{skillDir}}/sogni-gen.mjs --json --list-media all
 ```
 
+Do NOT use shell commands (`ls`, `cp`, etc.) to browse user media directories.
+
 ## Example Conversations
 
 User: "Draw a sunset over mountains"

package/mcp-server.mjs

@@ -580,6 +580,58 @@ The face likeness is preserved while applying the style from the prompt.`,
       properties: {},
     },
   },
+  {
+    name: 'extract_last_frame',
+    description: 'Extract the last frame from a video file as an image. Safe ffmpeg wrapper with input sanitization.',
+    inputSchema: {
+      type: 'object',
+      properties: {
+        video_path: {
+          type: 'string',
+          description: 'Path to the source video file',
+        },
+        output_path: {
+          type: 'string',
+          description: 'Path to save the extracted frame image (e.g. /tmp/lastframe.png)',
+        },
+      },
+      required: ['video_path', 'output_path'],
+    },
+  },
+  {
+    name: 'concat_videos',
+    description: 'Concatenate multiple video clips into a single video file. Safe ffmpeg wrapper with input sanitization. Requires at least 2 clips.',
+    inputSchema: {
+      type: 'object',
+      properties: {
+        output_path: {
+          type: 'string',
+          description: 'Path for the concatenated output video',
+        },
+        clips: {
+          type: 'array',
+          items: { type: 'string' },
+          description: 'Array of video clip file paths to concatenate (minimum 2)',
+          minItems: 2,
+        },
+      },
+      required: ['output_path', 'clips'],
+    },
+  },
+  {
+    name: 'list_media',
+    description: 'List recent inbound media files (images, audio, or all) from the user media directory (~/.clawdbot/media/inbound/). Returns the 5 most recent files sorted by modification time.',
+    inputSchema: {
+      type: 'object',
+      properties: {
+        type: {
+          type: 'string',
+          enum: ['images', 'audio', 'all'],
+          description: 'Type of media to list (default: images)',
+        },
+      },
+    },
+  },
 ];
 
 // ---------------------------------------------------------------------------
@@ -697,6 +749,40 @@ Defaults:
   return { content: [{ type: 'text', text }] };
 }
 
+async function handleExtractLastFrame(params) {
+  const videoPath = sanitizeString(params.video_path, 'video_path');
+  const outputPath = sanitizeString(params.output_path, 'output_path');
+  const result = await runSogniGen(['--extract-last-frame', videoPath, outputPath], { timeoutMs: 30_000 });
+  if (result.success === false) return formatError(result);
+  return { content: [{ type: 'text', text: `Extracted last frame to: ${result.outputPath || outputPath}` }] };
+}
+
+async function handleConcatVideos(params) {
+  const outputPath = sanitizeString(params.output_path, 'output_path');
+  if (!params.clips || params.clips.length < 2) {
+    return { content: [{ type: 'text', text: 'Error: At least 2 clips are required.' }], isError: true };
+  }
+  const clips = params.clips.map((c, i) => sanitizeString(c, `clips[${i}]`));
+  const result = await runSogniGen(['--concat-videos', outputPath, ...clips], { timeoutMs: 60_000 });
+  if (result.success === false) return formatError(result);
+  return { content: [{ type: 'text', text: `Concatenated ${result.clipCount || clips.length} clips to: ${result.outputPath || outputPath}` }] };
+}
+
+async function handleListMedia(params) {
+  const args = ['--list-media'];
+  if (params.type) {
+    args.push(validateEnum(params.type, ['images', 'audio', 'all'], 'type'));
+  }
+  const result = await runSogniGen(args, { timeoutMs: 10_000 });
+  if (result.success === false) return formatError(result);
+  const files = result.files || [];
+  if (files.length === 0) {
+    return { content: [{ type: 'text', text: `No ${result.mediaType || 'media'} files found.` }] };
+  }
+  const lines = files.map(f => `${f.name}  (${f.size} bytes, ${f.modified})\n  ${f.path}`);
+  return { content: [{ type: 'text', text: `Recent ${result.mediaType || 'media'} (${files.length}):\n${lines.join('\n')}` }] };
+}
+
 // ---------------------------------------------------------------------------
 // Server setup
 // ---------------------------------------------------------------------------
@@ -726,6 +812,12 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
         return handleListModels();
       case 'get_version':
         return await handleGetVersion();
+      case 'extract_last_frame':
+        return await handleExtractLastFrame(params);
+      case 'concat_videos':
+        return await handleConcatVideos(params);
+      case 'list_media':
+        return await handleListMedia(params);
       default:
         return {
           content: [{ type: 'text', text: `Unknown tool: ${name}` }],

package/openclaw.plugin.json

@@ -2,7 +2,7 @@
   "id": "sogni-gen",
   "name": "Sogni Image & Video Generation",
   "description": "Generate images and videos using Sogni AI via the sogni-gen skill.",
-  "version": "1.5.2",
+  "version": "1.5.3",
   "skills": [
     "."
   ],

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "sogni-gen",
-  "version": "1.5.2",
+  "version": "1.5.3",
   "description": "Sogni AI image & video generation — OpenClaw plugin and MCP server for Claude Code / Claude Desktop",
   "type": "module",
   "main": "sogni-gen.mjs",

package/sogni-gen.mjs

@@ -8,7 +8,7 @@ import { SogniClientWrapper, ClientEvent, getMaxContextImages } from '@sogni-ai/
 import JSON5 from 'json5';
 import { createHash, randomBytes } from 'crypto';
 import { spawnSync } from 'child_process';
-import { readFileSync, writeFileSync, existsSync, mkdirSync, mkdtempSync, statSync } from 'fs';
+import { readFileSync, writeFileSync, existsSync, mkdirSync, mkdtempSync, statSync, readdirSync, realpathSync, lstatSync } from 'fs';
 import { join, dirname, basename, extname } from 'path';
 import { homedir, tmpdir } from 'os';
 import sharp from 'sharp';
@@ -696,7 +696,12 @@ const options = {
   sam2Coordinates: null, // SAM2 coordinates for animate-replace [{x,y}]
   trimEndFrame: false, // Trim last frame for seamless stitching
   firstFrameStrength: null, // Keyframe interpolation (0.0-1.0)
-  lastFrameStrength: null // Keyframe interpolation (0.0-1.0)
+  lastFrameStrength: null, // Keyframe interpolation (0.0-1.0)
+  extractLastFrame: null, // --extract-last-frame <video> <image>
+  extractLastFrameOutput: null,
+  concatVideos: null, // --concat-videos <out> <clip1> <clip2> [...]
+  concatVideosClips: null,
+  listMedia: null // --list-media [images|audio|all]
 };
 const cliSet = {
   output: false,
@@ -993,6 +998,39 @@ for (let i = 0; i < args.length; i++) {
     i++;
     options.lastFrameStrength = parseNumberValue(raw, arg);
     cliSet.lastFrameStrength = true;
+  } else if (arg === '--extract-last-frame') {
+    const videoArg = requireFlagValue(args, i, arg);
+    i++;
+    const imageArg = requireFlagValue(args, i, arg + ' (output image)');
+    i++;
+    options.extractLastFrame = videoArg;
+    options.extractLastFrameOutput = imageArg;
+  } else if (arg === '--concat-videos') {
+    // Consume remaining positional args: <output> <clip1> <clip2> [clip3...]
+    const outArg = requireFlagValue(args, i, arg + ' (output path)');
+    i++;
+    const clips = [];
+    while (i + 1 < args.length && !args[i + 1].startsWith('-')) {
+      i++;
+      clips.push(args[i]);
+    }
+    if (clips.length < 2) {
+      fatalCliError('--concat-videos requires at least 2 clip paths after the output path.', {
+        code: 'INVALID_ARGUMENT',
+        details: { flag: '--concat-videos', clipsProvided: clips.length }
+      });
+    }
+    options.concatVideos = outArg;
+    options.concatVideosClips = clips;
+  } else if (arg === '--list-media') {
+    // Optional type argument (images|audio|all), default: images
+    const next = args[i + 1];
+    if (next && !next.startsWith('-') && ['images', 'audio', 'all'].includes(next)) {
+      i++;
+      options.listMedia = next;
+    } else {
+      options.listMedia = 'images';
+    }
   } else if (arg === '--last-image') {
     // Use image from last render as reference/context
     if (existsSync(LAST_RENDER_PATH)) {
@@ -1098,6 +1136,9 @@ General:
   --token-type <type>   Token type: spark|sogni (default: spark)
   --balance, --balances Show SPARK/SOGNI balances and exit
   --version, -V         Show sogni-gen version and exit
+  --extract-last-frame <video> <image>  Extract last frame from a video (safe ffmpeg wrapper)
+  --concat-videos <out> <clips...>      Concatenate video clips (safe ffmpeg wrapper, min 2 clips)
+  --list-media [type]   List recent inbound media files (images|audio|all, default: images)
   --last                Show last render info (JSON)
   --json                Output JSON with all details
   --strict-size         Do not auto-adjust video size to satisfy i2v reference resizing constraints
@@ -1440,7 +1481,7 @@ if (options.video) {
   options.model = options.model || openclawConfig?.defaultImageModel || 'z_image_turbo_bf16';
 }
 
-if (!options.prompt && !options.estimateVideoCost && !options.multiAngle && !options.showBalance && !options.showVersion) {
+if (!options.prompt && !options.estimateVideoCost && !options.multiAngle && !options.showBalance && !options.showVersion && !options.extractLastFrame && !options.concatVideos && !options.listMedia) {
   fatalCliError('No prompt provided. Use --help for usage.', { code: 'INVALID_ARGUMENT' });
 }
 
@@ -1717,7 +1758,7 @@ if (options.lastSeed) {
   }
 }
 
-if (!options.estimateVideoCost && !options.showVersion && (options.seed === null || options.seed === undefined)) {
+if (!options.estimateVideoCost && !options.showVersion && !options.extractLastFrame && !options.concatVideos && !options.listMedia && (options.seed === null || options.seed === undefined)) {
   const strategy = options.seedStrategy || openclawConfig?.seedStrategy || 'prompt-hash';
   const normalized = normalizeSeedStrategy(strategy) || 'prompt-hash';
   options.seedStrategy = normalized;
@@ -2459,6 +2500,120 @@ async function main() {
       return;
     }
 
+    // --- Utility commands (no Sogni auth required) ---
+
+    if (options.extractLastFrame) {
+      const videoPath = sanitizePath(options.extractLastFrame, '--extract-last-frame video');
+      const outputPath = sanitizePath(options.extractLastFrameOutput, '--extract-last-frame output');
+      if (!existsSync(videoPath)) {
+        const err = new Error(`Video file not found: ${videoPath}`);
+        err.code = 'FILE_NOT_FOUND';
+        throw err;
+      }
+      extractLastFrameFromVideo(videoPath, outputPath);
+      if (options.json || JSON_ERROR_MODE) {
+        console.log(JSON.stringify({
+          success: true,
+          type: 'extract-last-frame',
+          outputPath,
+          timestamp: new Date().toISOString()
+        }));
+      } else {
+        console.log(`Extracted last frame to: ${outputPath}`);
+      }
+      return;
+    }
+
+    if (options.concatVideos) {
+      const outputPath = sanitizePath(options.concatVideos, '--concat-videos output');
+      const clips = options.concatVideosClips.map((c, i) => sanitizePath(c, `clip[${i}]`));
+      for (const clip of clips) {
+        if (!existsSync(clip)) {
+          const err = new Error(`Clip file not found: ${clip}`);
+          err.code = 'FILE_NOT_FOUND';
+          throw err;
+        }
+      }
+      buildConcatVideoFromClips(outputPath, clips);
+      if (options.json || JSON_ERROR_MODE) {
+        console.log(JSON.stringify({
+          success: true,
+          type: 'concat-videos',
+          outputPath,
+          clipCount: clips.length,
+          timestamp: new Date().toISOString()
+        }));
+      } else {
+        console.log(`Concatenated ${clips.length} clips to: ${outputPath}`);
+      }
+      return;
+    }
+
+    if (options.listMedia) {
+      const mediaType = options.listMedia;
+      const baseDir = join(homedir(), '.clawdbot', 'media', 'inbound');
+
+      const IMAGE_EXTS = new Set(['.jpg', '.jpeg', '.png', '.webp', '.gif']);
+      const AUDIO_EXTS = new Set(['.m4a', '.mp3', '.wav', '.ogg']);
+
+      let allowedExts;
+      if (mediaType === 'images') allowedExts = IMAGE_EXTS;
+      else if (mediaType === 'audio') allowedExts = AUDIO_EXTS;
+      else allowedExts = new Set([...IMAGE_EXTS, ...AUDIO_EXTS]);
+
+      const files = [];
+      if (existsSync(baseDir)) {
+        // Validate the base directory itself isn't a symlink pointing outside ~/.clawdbot/
+        const clawdbotRoot = join(homedir(), '.clawdbot');
+        const resolvedBase = realpathSync(baseDir);
+        if (!resolvedBase.startsWith(clawdbotRoot)) {
+          const err = new Error('Media directory resolves outside of ~/.clawdbot/');
+          err.code = 'INVALID_PATH';
+          throw err;
+        }
+
+        const entries = readdirSync(baseDir);
+        for (const entry of entries) {
+          const ext = extname(entry).toLowerCase();
+          if (!allowedExts.has(ext)) continue;
+          const fullPath = join(baseDir, entry);
+          // Skip symlinks
+          const lstats = lstatSync(fullPath);
+          if (lstats.isSymbolicLink()) continue;
+          if (!lstats.isFile()) continue;
+          files.push({
+            path: fullPath,
+            name: entry,
+            size: lstats.size,
+            modified: lstats.mtime.toISOString()
+          });
+        }
+        // Sort by mtime descending, return top 5
+        files.sort((a, b) => b.modified.localeCompare(a.modified));
+        files.splice(5);
+      }
+
+      if (options.json || JSON_ERROR_MODE) {
+        console.log(JSON.stringify({
+          success: true,
+          type: 'list-media',
+          mediaType,
+          files,
+          timestamp: new Date().toISOString()
+        }));
+      } else {
+        if (files.length === 0) {
+          console.log(`No ${mediaType} files found in ${baseDir}`);
+        } else {
+          console.log(`Recent ${mediaType} (${files.length}):`);
+          for (const f of files) {
+            console.log(`  ${f.name}  (${f.size} bytes, ${f.modified})`);
+          }
+        }
+      }
+      return;
+    }
+
     const creds = loadCredentials();
     log('Connecting to Sogni...');
     client = new SogniClientWrapper({