{
  "id": "TApiWf49Discourse01",
  "name": "Discourse Analysis Pipeline (TranscriptAPI)",
  "nodes": [
    {
      "parameters": {},
      "id": "49494949-4949-4949-8949-494949494901",
      "name": "Start",
      "type": "n8n-nodes-base.manualTrigger",
      "typeVersion": 1,
      "position": [
        -300,
        0
      ]
    },
    {
      "parameters": {
        "assignments": {
          "assignments": [
            {
              "id": "49494949-4949-4949-8949-4949494949a1",
              "name": "searchQuery",
              "value": "REPLACE_WITH_TOPIC_OR_QUERY",
              "type": "string"
            }
          ]
        },
        "options": {}
      },
      "id": "49494949-4949-4949-8949-494949494902",
      "name": "Set Topic",
      "type": "n8n-nodes-base.set",
      "typeVersion": 3.4,
      "position": [
        -80,
        0
      ]
    },
    {
      "parameters": {
        "url": "https://transcriptapi.com/api/v2/youtube/search",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth",
        "sendQuery": true,
        "queryParameters": {
          "parameters": [
            {
              "name": "q",
              "value": "={{ $json.searchQuery }}"
            },
            {
              "name": "type",
              "value": "video"
            }
          ]
        },
        "options": {}
      },
      "id": "49494949-4949-4949-8949-494949494903",
      "name": "Search Videos (TranscriptAPI)",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        140,
        0
      ],
      "credentials": {
        "httpHeaderAuth": {
          "id": "REPLACE_TRANSCRIPTAPI_CRED_ID",
          "name": "TranscriptAPI - Authorization Bearer"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "const data = $input.first().json;\nreturn (data.results || []).filter((r) => r.hasCaptions).map((r) => ({ json: { videoId: r.videoId, title: r.title || null, channelTitle: r.channelTitle || null, videoUrl: 'https://www.youtube.com/watch?v=' + r.videoId } }));"
      },
      "id": "49494949-4949-4949-8949-494949494904",
      "name": "Extract Captioned Results",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        360,
        0
      ]
    },
    {
      "parameters": {
        "maxItems": 5
      },
      "id": "49494949-4949-4949-8949-494949494905",
      "name": "Keep Top 5",
      "type": "n8n-nodes-base.limit",
      "typeVersion": 1,
      "position": [
        580,
        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": "49494949-4949-4949-8949-494949494906",
      "name": "Get Transcript (TranscriptAPI)",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        800,
        0
      ],
      "credentials": {
        "httpHeaderAuth": {
          "id": "REPLACE_TRANSCRIPTAPI_CRED_ID",
          "name": "TranscriptAPI - Authorization Bearer"
        }
      },
      "onError": "continueRegularOutput"
    },
    {
      "parameters": {
        "jsCode": "function ts(sec) {\n  const s = Math.floor(sec || 0);\n  const h = Math.floor(s / 3600), m = Math.floor((s % 3600) / 60), r = s % 60;\n  const mm = (r < 10 ? '0' : '') + r;\n  return h > 0 ? (h + ':' + (m < 10 ? '0' : '') + m + ':' + mm) : (m + ':' + mm);\n}\nconst MAX = 4000; const sources = []; const blocks = [];\nfor (const item of $input.all()) {\n  const d = item.json; const seg = Array.isArray(d.transcript) ? d.transcript : [];\n  if (seg.length === 0) continue;\n  const meta = d.metadata || {}; const vid = d.video_id; const title = meta.title || vid;\n  const url = 'https://www.youtube.com/watch?v=' + vid;\n  const lines = seg.map((s) => '[' + ts(s.start) + '] ' + s.text).join('\\n').slice(0, MAX);\n  sources.push({ title, videoUrl: url });\n  blocks.push('### ' + title + ' (' + url + ')\\n' + lines);\n}\nreturn [ { json: { topic: $('Set Topic').first().json.searchQuery, sources, sourceCount: sources.length, combinedText: blocks.join('\\n\\n') } } ];"
      },
      "id": "49494949-4949-4949-8949-494949494907",
      "name": "Collect Transcripts",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        1020,
        0
      ]
    },
    {
      "parameters": {
        "method": "POST",
        "url": "https://api.openai.com/v1/chat/completions",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth",
        "sendBody": true,
        "specifyBody": "json",
        "jsonBody": "={{ JSON.stringify({ model: 'gpt-4o-mini', response_format: { type: 'json_object' }, messages: [ { role: 'system', content: 'You are a neutral discourse analyst. Using ONLY the supplied timestamped transcripts about a topic, produce a thematic analysis. Respond with a JSON object with keys: recurring_frames (array of strings — common ways the topic is framed), stance_clusters (array of objects with stance (a short label) and summary), evidence_snippets (array of objects with quote (short, near-verbatim), timestamp (the [m:ss] marker), source_title), caveats (array of limitations of this analysis). Stay descriptive and neutral; do not take sides or invent content.' }, { role: 'user', content: 'Topic: ' + $json.topic + '\\n\\nSource transcripts (timestamped):\\n\\n' + $json.combinedText } ] }) }}",
        "options": {}
      },
      "id": "49494949-4949-4949-8949-494949494908",
      "name": "Discourse Analysis (LLM)",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        1240,
        0
      ],
      "credentials": {
        "httpHeaderAuth": {
          "id": "REPLACE_OPENAI_CRED_ID",
          "name": "OpenAI - Authorization Bearer"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "const raw = ($json.choices && $json.choices[0] && $json.choices[0].message.content) || '';\nlet r; try { r = JSON.parse(raw); } catch (e) { r = {}; }\nconst src = $('Collect Transcripts').first().json;\nreturn [ { json: { topic: src.topic, recurringFrames: r.recurring_frames || [], stanceClusters: r.stance_clusters || [], evidenceSnippets: r.evidence_snippets || [], caveats: r.caveats || [], sources: src.sources } } ];"
      },
      "id": "49494949-4949-4949-8949-494949494909",
      "name": "Format Report",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        1460,
        0
      ]
    }
  ],
  "connections": {
    "Start": {
      "main": [
        [
          {
            "node": "Set Topic",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Set Topic": {
      "main": [
        [
          {
            "node": "Search Videos (TranscriptAPI)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Search Videos (TranscriptAPI)": {
      "main": [
        [
          {
            "node": "Extract Captioned Results",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Extract Captioned Results": {
      "main": [
        [
          {
            "node": "Keep Top 5",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Keep Top 5": {
      "main": [
        [
          {
            "node": "Get Transcript (TranscriptAPI)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Get Transcript (TranscriptAPI)": {
      "main": [
        [
          {
            "node": "Collect Transcripts",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Collect Transcripts": {
      "main": [
        [
          {
            "node": "Discourse Analysis (LLM)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Discourse Analysis (LLM)": {
      "main": [
        [
          {
            "node": "Format Report",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "active": false,
  "settings": {
    "executionOrder": "v1"
  },
  "pinData": {}
}