{
  "id": "TApiWf04Rag01",
  "name": "Playlist/Channel → Transcripts → RAG Chunks (TranscriptAPI)",
  "nodes": [
    {
      "parameters": {},
      "id": "44444444-4444-4444-8444-444444444401",
      "name": "Start",
      "type": "n8n-nodes-base.manualTrigger",
      "typeVersion": 1,
      "position": [
        -200,
        0
      ]
    },
    {
      "parameters": {
        "assignments": {
          "assignments": [
            {
              "id": "a4444444-4444-4444-8444-444444444aaa",
              "name": "playlistUrl",
              "value": "REPLACE_WITH_PLAYLIST_URL_OR_ID",
              "type": "string"
            }
          ]
        },
        "options": {}
      },
      "id": "44444444-4444-4444-8444-444444444402",
      "name": "Set Playlist Source",
      "type": "n8n-nodes-base.set",
      "typeVersion": 3.4,
      "position": [
        20,
        0
      ]
    },
    {
      "parameters": {
        "url": "https://transcriptapi.com/api/v2/youtube/playlist/videos",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth",
        "sendQuery": true,
        "queryParameters": {
          "parameters": [
            {
              "name": "playlist",
              "value": "={{ $json.playlistUrl }}"
            }
          ]
        },
        "options": {}
      },
      "id": "44444444-4444-4444-8444-444444444403",
      "name": "Get Playlist Videos (TranscriptAPI)",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        240,
        0
      ],
      "credentials": {
        "httpHeaderAuth": {
          "id": "REPLACE_TRANSCRIPTAPI_CRED_ID",
          "name": "TranscriptAPI - Authorization Bearer"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "// Split the playlist/videos response into one item per video (watch URL + title).\nconst data = $input.first().json;\nconst videos = data.results || [];\nreturn videos.map((v) => ({\n  json: {\n    videoId: v.videoId,\n    videoTitle: v.title || null,\n    videoUrl: 'https://www.youtube.com/watch?v=' + v.videoId,\n  },\n}));"
      },
      "id": "44444444-4444-4444-8444-444444444404",
      "name": "Extract Video List",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        460,
        0
      ]
    },
    {
      "parameters": {
        "maxItems": 5
      },
      "id": "44444444-4444-4444-8444-444444444405",
      "name": "Keep First 5",
      "type": "n8n-nodes-base.limit",
      "typeVersion": 1,
      "position": [
        680,
        0
      ]
    },
    {
      "parameters": {
        "url": "https://transcriptapi.com/api/v2/youtube/transcript",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth",
        "sendQuery": true,
        "queryParameters": {
          "parameters": [
            {
              "name": "video_url",
              "value": "={{ $json.videoUrl }}"
            },
            {
              "name": "format",
              "value": "json"
            },
            {
              "name": "send_metadata",
              "value": "true"
            }
          ]
        },
        "options": {}
      },
      "id": "44444444-4444-4444-8444-444444444406",
      "name": "Get Transcript (TranscriptAPI)",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        900,
        0
      ],
      "credentials": {
        "httpHeaderAuth": {
          "id": "REPLACE_TRANSCRIPTAPI_CRED_ID",
          "name": "TranscriptAPI - Authorization Bearer"
        }
      },
      "onError": "continueRegularOutput"
    },
    {
      "parameters": {
        "jsCode": "// Turn each video transcript into clean, overlapping text chunks ready to embed.\n// Output: one item per chunk, shaped for a vector DB ({ id, text, metadata }).\nconst CHUNK_WORDS = 200;   // embedding-friendly window size\nconst OVERLAP_WORDS = 30;  // context overlap between adjacent chunks\nconst out = [];\nfor (const item of $input.all()) {\n  const d = item.json;\n  const segments = Array.isArray(d.transcript) ? d.transcript : [];\n  if (segments.length === 0) continue; // skip videos without captions (cost 0 credits)\n  const meta = d.metadata || {};\n  const videoId = d.video_id;\n  const words = segments.map((s) => s.text).join(' ').split(' ').filter((w) => w.length > 0);\n  const step = Math.max(1, CHUNK_WORDS - OVERLAP_WORDS);\n  const chunks = [];\n  for (let i = 0; i < words.length; i += step) {\n    chunks.push(words.slice(i, i + CHUNK_WORDS).join(' '));\n    if (i + CHUNK_WORDS >= words.length) break;\n  }\n  chunks.forEach((text, idx) => {\n    out.push({\n      json: {\n        id: videoId + '-' + idx,\n        text,\n        metadata: {\n          video_id: videoId,\n          video_title: meta.title || null,\n          video_url: 'https://www.youtube.com/watch?v=' + videoId,\n          author: meta.author_name || null,\n          language: d.language || null,\n          chunk_index: idx,\n          total_chunks: chunks.length,\n          char_count: text.length,\n        },\n      },\n    });\n  });\n}\nreturn out;"
      },
      "id": "44444444-4444-4444-8444-444444444407",
      "name": "Chunk for RAG",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        1120,
        0
      ]
    }
  ],
  "connections": {
    "Start": {
      "main": [
        [
          {
            "node": "Set Playlist Source",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Set Playlist Source": {
      "main": [
        [
          {
            "node": "Get Playlist Videos (TranscriptAPI)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Get Playlist Videos (TranscriptAPI)": {
      "main": [
        [
          {
            "node": "Extract Video List",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Extract Video List": {
      "main": [
        [
          {
            "node": "Keep First 5",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Keep First 5": {
      "main": [
        [
          {
            "node": "Get Transcript (TranscriptAPI)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Get Transcript (TranscriptAPI)": {
      "main": [
        [
          {
            "node": "Chunk for RAG",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "active": false,
  "settings": {
    "executionOrder": "v1"
  },
  "pinData": {}
}