Skip to main content
Glama

play_shorts

Stream YouTube Shorts as an auto-advancing playlist from specific channels or subscriptions, automatically loading more content as you watch.

Instructions

Play Shorts as a continuous auto-advancing playlist. Automatically fetches more as you watch. Fetch from a specific channel or from your subscribed channels.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
sourceYes"channel" or "subscriptions"
channel_urlNoRequired when source is "channel". YouTube channel URL.
max_channelsNoWhen source is "subscriptions": channels to sample (default 15)
shorts_per_channelNoWhen source is "subscriptions": shorts per channel (default 3)
limitNoWhen source is "channel": max shorts (default 15)
shuffleNoShuffle the playback order

Implementation Reference

  • Tool registration and handler implementation for 'play_shorts' which plays YouTube Shorts in mpv, supports channel or subscription sources, and includes auto-refill for channel sources.
    server.tool(
      'play_shorts',
      'Play Shorts as a continuous auto-advancing playlist. Automatically fetches more as you watch. Fetch from a specific channel or from your subscribed channels.',
      {
        source: z.enum(['channel', 'subscriptions']).describe('"channel" or "subscriptions"'),
        channel_url: z.string().url().optional().describe('Required when source is "channel". YouTube channel URL.'),
        max_channels: z.number().min(1).max(50).default(15).describe('When source is "subscriptions": channels to sample (default 15)'),
        shorts_per_channel: z.number().min(1).max(10).default(3).describe('When source is "subscriptions": shorts per channel (default 3)'),
        limit: z.number().min(1).max(50).default(15).describe('When source is "channel": max shorts (default 15)'),
        shuffle: z.boolean().default(false).describe('Shuffle the playback order'),
      },
      async ({ source, channel_url, max_channels, shorts_per_channel, limit, shuffle }) => {
        const depErr = checkDeps();
        if (depErr) return errorResult(depErr);
    
        if (source === 'channel') {
          if (!channel_url) return errorResult('channel_url is required when source is "channel".');
          const urlErr = validateYouTubeUrl(channel_url);
          if (urlErr) return errorResult(urlErr);
        }
    
        let urls: string[];
        try {
          if (source === 'channel') {
            const result = await fetchFeed(`${stripChannelSuffix(channel_url!)}/shorts`, limit);
            urls = (result.entries || []).map((e) => e.url as string).filter(Boolean);
          } else {
            const sub = await fetchSubscriptionShortUrls(max_channels, shorts_per_channel);
            urls = sub.urls;
          }
        } catch (err) {
          return errorResult(`Error fetching shorts: ${err instanceof Error ? err.message : String(err)}`);
        }
    
        if (urls.length === 0) return textResult('No shorts found.');
    
        const playlistFile = mpv.writeTempPlaylist(urls);
    
        try {
          await mpv.launch({ playlistFile, shuffle, socketTimeoutMs: 15_000 });
        } catch {
          return errorResult('mpv failed to start. Run `mpv <url>` manually to see the error.');
        }
    
        // Auto-refill for channel source (subscription source is harder to paginate)
        if (source === 'channel' && channel_url) {
          const shortsUrl = `${stripChannelSuffix(channel_url)}/shorts`;
          mpv.startAutoRefill(urls.length, async (offset, batch) => {
            const more = await fetchFeed(shortsUrl, batch, offset + 1);
            return (more.entries || []).map((e) => e.url as string).filter(Boolean);
          });
        }
    
        let title = 'Shorts playlist';
        try { title = (await mpv.getProperty('media-title')) as string || title; } catch { /* loading */ }
    
        return textResult({ status: 'playing_shorts', title, total: urls.length, shuffle, source, autoRefill: source === 'channel' });
      }
    );

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/ronantakizawa/social-video-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server