workflow 08: v1.1 fixes + docs update

- 08-class-recording-rag-v1.1.json: Whisper URL → 10.5.1.16:8786, model → faster-distil-whisper-small.en, auth headers → env vars, inputDataFieldName fix, Paperless response format text - PLAUSIBLEDENABILITY.md: remove Whisper from GPU list (moved to N.O.M.A.D.) - n8n-workflows/README.md: fix class recordings path to /mnt/tank, correct workflow 08 trigger (webhook, not folder watcher), add curl upload example and docker-compose setup instructions
2026-05-09 19:07:27 -05:00
parent 11904f39b2
commit 9aef43ee10
3 changed files with 567 additions and 16 deletions
--- a/n8n-workflows/08-class-recording-rag-v1.1.json
+++ b/n8n-workflows/08-class-recording-rag-v1.1.json
@@ -0,0 +1,538 @@
+{
+  "updatedAt": "2026-05-07T01:16:50.340Z",
+  "createdAt": "2026-05-07T01:16:50.340Z",
+  "id": "91I278E72FAaQenD",
+  "name": "Class Recording → RAG v1.1",
+  "description": null,
+  "active": true,
+  "isArchived": false,
+  "nodes": [
+    {
+      "parameters": {
+        "httpMethod": "POST",
+        "path": "class/upload",
+        "responseMode": "lastNode",
+        "responseData": "allEntries",
+        "options": {
+          "rawBody": true
+        }
+      },
+      "id": "e475f17b-8560-4376-852e-a613ba6b300e",
+      "name": "Class Upload Webhook",
+      "type": "n8n-nodes-base.webhook",
+      "typeVersion": 2,
+      "position": [
+        0,
+        0
+      ],
+      "webhookId": "d0ee8226-db56-477e-8a3f-572a9129aea9"
+    },
+    {
+      "parameters": {
+        "jsCode": "const data = $input.first().json;\nconst body = data.body || data;\n\n// Class name comes from:\n// 1. Form field 'class_name' in multipart upload\n// 2. Query parameter ?class=CS101\n// 3. Header X-Class-Name\nconst className = body.class_name || data.headers?.['x-class-name'] || data.query?.class || 'unclassified';\n\n// Lecture title from:\n// 1. Form field 'lecture_title'\n// 2. Query param ?title=...\n// 3. Header X-Lecture-Title\n// 4. Form field 'filename' (original filename)\nconst rawTitle = body.lecture_title || data.headers?.['x-lecture-title'] || data.query?.title || body.filename || 'Untitled Lecture';\nconst lectureTitle = rawTitle.replace(/\\.wav$/i, '').replace(/[-_]/g, ' ');\n\nconst classSlug = className.toLowerCase().replace(/[^a-z0-9]+/g, '-').replace(/^-|-$/g, '');\n\nreturn [{\n  json: {\n    className,\n    classSlug,\n    lectureTitle,\n    uploadedAt: new Date().toISOString()\n  }\n}];"
+      },
+      "id": "24e4a726-ab50-4767-a144-1751616c781d",
+      "name": "Extract Class Info",
+      "type": "n8n-nodes-base.code",
+      "typeVersion": 2,
+      "position": [
+        224,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "http://10.5.1.16:8786/v1/audio/transcriptions",
+        "sendHeaders": true,
+        "headerParameters": {
+          "parameters": [
+            {
+              "name": "Accept",
+              "value": "application/json"
+            }
+          ]
+        },
+        "sendBody": true,
+        "contentType": "multipart-form-data",
+        "bodyParameters": {
+          "parameters": [
+            {
+              "parameterType": "formBinaryData",
+              "name": "file",
+              "inputDataFieldName": "data"
+            },
+            {
+              "name": "model",
+              "value": "Systran/faster-distil-whisper-small.en"
+            },
+            {
+              "name": "language",
+              "value": "en"
+            },
+            {
+              "name": "response_format",
+              "value": "verbose_json"
+            }
+          ]
+        },
+        "options": {
+          "timeout": 600000
+        }
+      },
+      "id": "fcd1062d-8f60-4214-85ee-f53b914abcbb",
+      "name": "Whisper Transcribe",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        448,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "jsCode": "const result = $input.first().json;\nconst classInfo = $('Extract Class Info').first().json;\n\nconst segments = result.segments || [];\nconst fullText = result.text || segments.map(s => s.text).join(' ');\nconst duration = result.duration || segments[segments.length - 1]?.end || 0;\n\nconst timestamped = segments.map(s => {\n  const mins = Math.floor(s.start / 60);\n  const secs = Math.floor(s.start % 60);\n  return `[${mins}:${String(secs).padStart(2, '0')}] ${s.text.trim()}`;\n}).join('\\n');\n\nreturn [{\n  json: {\n    className: classInfo.className,\n    classSlug: classInfo.classSlug,\n    lectureTitle: classInfo.lectureTitle,\n    fullText: fullText.trim(),\n    timestampedText: timestamped,\n    durationMinutes: Math.round(duration / 60),\n    segmentCount: segments.length\n  }\n}];"
+      },
+      "id": "ecf09ad2-327e-4a82-90f3-95da5590e0ac",
+      "name": "Format Transcript",
+      "type": "n8n-nodes-base.code",
+      "typeVersion": 2,
+      "position": [
+        672,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "http://10.5.1.6:4000/v1/chat/completions",
+        "sendHeaders": true,
+        "headerParameters": {
+          "parameters": [
+            {
+              "name": "Content-Type",
+              "value": "application/json"
+            },
+            {
+              "name": "Authorization",
+              "value": "=Bearer {{ $env.LITELLM_API_KEY }}"
+            }
+          ]
+        },
+        "sendBody": true,
+        "specifyBody": "json",
+        "jsonBody": "={\n  \"model\": \"medium\",\n  \"messages\": [\n    {\n      \"role\": \"system\",\n      \"content\": \"You are an academic note-taking assistant. Given a class lecture transcription, provide:\\n1. A concise summary (3-5 sentences)\\n2. Key concepts covered (as a list)\\n3. Important definitions or formulas\\n4. Any assignments or deadlines mentioned\\n\\nRespond in JSON with keys: summary, key_concepts (array), definitions (array of {term, definition}), assignments (array)\"\n    },\n    {\n      \"role\": \"user\",\n      \"content\": \"Class: {{ $json.className }}\\nLecture: {{ $json.lectureTitle }}\\n\\nTranscription (first 3000 chars):\\n{{ $json.fullText.substring(0, 3000) }}\"\n    }\n  ],\n  \"max_tokens\": 600,\n  \"temperature\": 0.2,\n  \"response_format\": { \"type\": \"json_object\" }\n}",
+        "options": {}
+      },
+      "id": "e7407d06-f7f4-4105-a817-055d4703d59d",
+      "name": "AI Extract Notes",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        880,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "jsCode": "const llmResponse = $input.first().json;\nconst transcriptData = $('Format Transcript').first().json;\n\nlet notes;\ntry {\n  notes = JSON.parse(llmResponse.choices?.[0]?.message?.content || '{}');\n} catch(e) {\n  notes = { summary: 'Could not parse notes', key_concepts: [], definitions: [], assignments: [] };\n}\n\nconst ragText = [\n  `Class: ${transcriptData.className}`,\n  `Lecture: ${transcriptData.lectureTitle}`,\n  `Duration: ${transcriptData.durationMinutes} minutes`,\n  `Summary: ${notes.summary}`,\n  `Key Concepts: ${(notes.key_concepts || []).join(', ')}`,\n  '',\n  'Full Transcript:',\n  transcriptData.fullText\n].join('\\n');\n\nreturn [{\n  json: {\n    className: transcriptData.className,\n    classSlug: transcriptData.classSlug,\n    lectureTitle: transcriptData.lectureTitle,\n    durationMinutes: transcriptData.durationMinutes,\n    notes,\n    ragText,\n    source: `class-${transcriptData.classSlug}-${Date.now()}`,\n    metadata: {\n      class_name: transcriptData.className,\n      class_slug: transcriptData.classSlug,\n      lecture_title: transcriptData.lectureTitle,\n      duration_minutes: transcriptData.durationMinutes,\n      type: 'class_recording',\n      recorded_at: new Date().toISOString()\n    }\n  }\n}];"
+      },
+      "id": "71b0807d-03ba-45b2-b75b-b6fbeee5687e",
+      "name": "Prepare for RAG",
+      "type": "n8n-nodes-base.code",
+      "typeVersion": 2,
+      "position": [
+        1104,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "jsCode": "const input = $input.first().json;\nconst text = input.ragText || '';\nconst source = input.source;\nconst metadata = input.metadata;\n\nconst CHUNK_SIZE = 500;\nconst OVERLAP = 50;\n\nconst chunks = [];\nconst sentences = text.split(/(?<=[.!?])\\s+/);\nlet current = '';\n\nfor (const sentence of sentences) {\n  if ((current + ' ' + sentence).length > CHUNK_SIZE && current.length > 0) {\n    chunks.push(current.trim());\n    const words = current.split(' ');\n    current = words.slice(-OVERLAP).join(' ') + ' ' + sentence;\n  } else {\n    current = current ? current + ' ' + sentence : sentence;\n  }\n}\nif (current.trim()) chunks.push(current.trim());\n\nreturn chunks.map((chunk, i) => ({\n  json: {\n    chunk,\n    chunkIndex: i,\n    totalChunks: chunks.length,\n    source,\n    metadata\n  }\n}));"
+      },
+      "id": "9e4bbe70-72b0-4955-848f-0d1b62ae94a4",
+      "name": "Chunk Text",
+      "type": "n8n-nodes-base.code",
+      "typeVersion": 2,
+      "position": [
+        1328,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "http://10.5.1.6:11434/v1/embeddings",
+        "sendHeaders": true,
+        "headerParameters": {
+          "parameters": [
+            {
+              "name": "Content-Type",
+              "value": "application/json"
+            }
+          ]
+        },
+        "sendBody": true,
+        "specifyBody": "json",
+        "jsonBody": "={\n  \"model\": \"nomic-embed-text\",\n  \"input\": \"{{ $json.chunk }}\"\n}",
+        "options": {
+          "batching": {
+            "batch": {
+              "batchSize": 5,
+              "batchInterval": 500
+            }
+          }
+        }
+      },
+      "id": "4ae1ab11-c76f-4478-84e0-7bcdfb20b83b",
+      "name": "Get Embedding",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        1552,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "jsCode": "const embeddingResponse = $input.first().json;\nconst chunkData = $('Chunk Text').item;\nconst embedding = embeddingResponse.data?.[0]?.embedding || [];\n\nconst pointId = 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx'.replace(/[xy]/g, c => {\n  const r = Math.random() * 16 | 0;\n  return (c === 'x' ? r : (r & 0x3 | 0x8)).toString(16);\n});\n\nreturn [{\n  json: {\n    id: pointId,\n    vector: embedding,\n    payload: {\n      text: chunkData.json.chunk,\n      source: chunkData.json.source,\n      chunkIndex: chunkData.json.chunkIndex,\n      totalChunks: chunkData.json.totalChunks,\n      metadata: chunkData.json.metadata,\n      ingested_at: new Date().toISOString()\n    }\n  }\n}];"
+      },
+      "id": "d4232b02-23fc-4cbd-ac2d-2fdbfbe9583a",
+      "name": "Prepare Qdrant Point",
+      "type": "n8n-nodes-base.code",
+      "typeVersion": 2,
+      "position": [
+        1760,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "method": "PUT",
+        "url": "http://10.5.1.6:6333/collections/knowledge_base/points",
+        "sendHeaders": true,
+        "headerParameters": {
+          "parameters": [
+            {
+              "name": "Content-Type",
+              "value": "application/json"
+            }
+          ]
+        },
+        "sendBody": true,
+        "specifyBody": "json",
+        "jsonBody": "={\n  \"points\": [\n    {\n      \"id\": \"{{ $json.id }}\",\n      \"vector\": {{ JSON.stringify($json.vector) }},\n      \"payload\": {{ JSON.stringify($json.payload) }}\n    }\n  ]\n}",
+        "options": {}
+      },
+      "id": "845a2aa0-2546-431d-96e8-b36875d1fe7d",
+      "name": "Upsert to Qdrant",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        1984,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "jsCode": "const rag = $('Prepare for RAG').first().json;\nreturn [{\n  json: {\n    title:    `🎓 ${rag.className} Lecture Processed`,\n    message:  `${rag.lectureTitle} (${rag.durationMinutes}min)\\nSummary: ${rag.notes.summary}\\nKey concepts: ${(rag.notes.key_concepts || []).slice(0, 5).join(', ')}`,\n    priority: 5\n  }\n}];"
+      },
+      "id": "fmt08-0001-0000-0000-000000000008",
+      "name": "Format Gotify Message",
+      "type": "n8n-nodes-base.code",
+      "typeVersion": 2,
+      "position": [
+        1968,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "message": "={{ $json.message }}",
+        "additionalFields": {
+          "priority": "={{ $json.priority }}",
+          "title": "={{ $json.title }}"
+        },
+        "options": {
+          "contentType": "text/plain"
+        }
+      },
+      "id": "5b31b660-c414-40d1-82a5-d68511203133",
+      "name": "Notify via Gotify",
+      "type": "n8n-nodes-base.gotify",
+      "typeVersion": 1,
+      "position": [
+        2208,
+        0
+      ],
+      "credentials": {
+        "gotifyApi": {
+          "id": "SETUP_REQUIRED",
+          "name": "Class Recordings"
+        }
+      }
+    },
+    {
+      "parameters": {
+        "jsCode": "const rag = $('Prepare for RAG').first().json;\nconst notes = rag.notes || {};\nconst concepts = (notes.key_concepts || []).map((c,i) => `${i+1}. ${c}`).join('\\n');\nconst defs = Object.entries(notes.definitions || {}).map(([k,v]) => `${k}: ${v}`).join('\\n');\nconst assignments = (notes.assignments || []).map((a,i) => `${i+1}. ${a}`).join('\\n');\nconst content = [\n  `Lecture Notes: ${rag.lectureTitle}`,\n  `Class: ${rag.className}`,\n  `Duration: ${rag.durationMinutes} minutes`,\n  `Date: ${new Date().toISOString()}`, '',\n  `Summary`, `=======`, notes.summary || '', '',\n  `Key Concepts`, `=============`, concepts,\n  ...(defs     ? ['', `Definitions`,    `===========`, defs]        : []),\n  ...(assignments ? ['', `Assignments`, `===========`, assignments] : []),\n].join('\\n');\nconst title = `${rag.className} — ${rag.lectureTitle}`;\nreturn [{ binary: { document: { data: Buffer.from(content,'utf-8').toString('base64'), mimeType:'text/plain', fileName:`${title}.txt` }}, json: { title, className: rag.className } }];\n"
+      },
+      "id": "95aea3a7-3b15-4196-9e4a-dc8f6ed8d49e",
+      "name": "Prepare Notes Doc",
+      "type": "n8n-nodes-base.code",
+      "typeVersion": 2,
+      "position": [
+        2208,
+        -200
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "https://paperless.paccoco.com/api/documents/post_document/",
+        "sendHeaders": true,
+        "headerParameters": {
+          "parameters": [
+            {
+              "name": "Authorization",
+              "value": "=Token {{ $env.PAPERLESS_API_TOKEN }}"
+            }
+          ]
+        },
+        "sendBody": true,
+        "contentType": "multipart-form-data",
+        "bodyParameters": {
+          "parameters": [
+            {
+              "parameterType": "formBinaryData",
+              "name": "document",
+              "inputDataFieldName": "document"
+            },
+            {
+              "name": "title",
+              "value": "={{ $json.title }}"
+            },
+            {
+              "name": "tags[]",
+              "value": "={{ $env.PAPERLESS_TAG_LECTURE_NOTES }}"
+            }
+          ]
+        },
+        "options": {
+          "response": {
+            "response": {
+              "responseFormat": "text"
+            }
+          }
+        }
+      },
+      "id": "c902a153-ab26-4518-af2a-585f0b72fe6c",
+      "name": "Save Notes to Paperless",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        2432,
+        -200
+      ],
+      "continueOnFail": true
+    }
+  ],
+  "connections": {
+    "Class Upload Webhook": {
+      "main": [
+        [
+          {
+            "node": "Extract Class Info",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Extract Class Info": {
+      "main": [
+        [
+          {
+            "node": "Whisper Transcribe",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Whisper Transcribe": {
+      "main": [
+        [
+          {
+            "node": "Format Transcript",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Format Transcript": {
+      "main": [
+        [
+          {
+            "node": "AI Extract Notes",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "AI Extract Notes": {
+      "main": [
+        [
+          {
+            "node": "Prepare for RAG",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Prepare for RAG": {
+      "main": [
+        [
+          {
+            "node": "Chunk Text",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Chunk Text": {
+      "main": [
+        [
+          {
+            "node": "Get Embedding",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Get Embedding": {
+      "main": [
+        [
+          {
+            "node": "Prepare Qdrant Point",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Prepare Qdrant Point": {
+      "main": [
+        [
+          {
+            "node": "Upsert to Qdrant",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Upsert to Qdrant": {
+      "main": [
+        [
+          {
+            "node": "Format Gotify Message",
+            "type": "main",
+            "index": 0
+          },
+          {
+            "node": "Prepare Notes Doc",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Format Gotify Message": {
+      "main": [
+        [
+          {
+            "node": "Notify via Gotify",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Prepare Notes Doc": {
+      "main": [
+        [
+          {
+            "node": "Save Notes to Paperless",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    }
+  },
+  "settings": {
+    "executionOrder": "v1",
+    "binaryMode": "separate"
+  },
+  "staticData": null,
+  "meta": null,
+  "pinData": {},
+  "versionId": "d64293a7-5537-4b01-a11b-6b838d6cacbf",
+  "activeVersionId": "d64293a7-5537-4b01-a11b-6b838d6cacbf",
+  "versionCounter": 11,
+  "triggerCount": 1,
+  "tags": [
+    {
+      "updatedAt": "2026-05-07T00:49:56.081Z",
+      "createdAt": "2026-05-07T00:49:56.081Z",
+      "id": "Q9lCYtCOuy9XmeyL",
+      "name": "school"
+    },
+    {
+      "updatedAt": "2026-05-06T22:39:59.716Z",
+      "createdAt": "2026-05-06T22:39:59.716Z",
+      "id": "tuPxDoKFqpVrrQoX",
+      "name": "whisper"
+    },
+    {
+      "updatedAt": "2026-05-06T22:39:33.284Z",
+      "createdAt": "2026-05-06T22:39:33.284Z",
+      "id": "7uuEyQrIlcuruZBk",
+      "name": "rag"
+    },
+    {
+      "updatedAt": "2026-05-06T22:38:51.678Z",
+      "createdAt": "2026-05-06T22:38:51.678Z",
+      "id": "S9FxzVfHLsHmyzyt",
+      "name": "ai"
+    }
+  ],
+  "shared": [
+    {
+      "updatedAt": "2026-05-07T01:16:50.340Z",
+      "createdAt": "2026-05-07T01:16:50.340Z",
+      "role": "workflow:owner",
+      "workflowId": "91I278E72FAaQenD",
+      "projectId": "dxCRnBdX5uJizCGa",
+      "project": {
+        "updatedAt": "2026-05-06T01:10:37.638Z",
+        "createdAt": "2026-05-06T01:08:07.820Z",
+        "id": "dxCRnBdX5uJizCGa",
+        "name": "Wilfred Fizzlepoof <admin@paccoco.com>",
+        "type": "personal",
+        "icon": null,
+        "description": null,
+        "creatorId": "47b2227f-2fc2-4a08-bd35-ea157b92df0d"
+      }
+    }
+  ],
+  "versionMetadata": {
+    "name": "Version d64293a7",
+    "description": ""
+  }
+}
--- a/n8n-workflows/README.md
+++ b/n8n-workflows/README.md
@@ -11,7 +11,7 @@
 | 05 | Paperless → RAG | `POST /webhook/paperless/rag-ingest` | Auto-ingests Paperless documents into Qdrant for RAG search; notifies via Gotify |
 | 06 | RSS Digest | Scheduled (every 6h) + `POST /webhook/scrape/summarize` | Fetches RSS feeds, deduplicates via Postgres, AI-summarizes per-feed, pushes to Gotify |
 | 07 | Git Commit Summarizer | `POST /webhook/git/push` | Receives git push webhooks, AI-summarizes changes, notifies via Gotify |
-| 08 | Class Recording → RAG | Watches `/mnt/data/class-recordings/` | Auto-transcribes new .wav files, extracts key notes, ingests into RAG by class, saves notes to Paperless, notifies via Gotify |
+| 08 | Class Recording → RAG | `POST /webhook/class/upload` | Transcribes uploaded .wav, extracts key notes, ingests into RAG by class, saves notes to Paperless, notifies via Gotify |
 | 14 | Paperless Inbox Reminder | Scheduled (Monday 9 AM) | Queries Paperless for untagged documents; sends Gotify reminder with oldest items listed, or a ✅ clear if inbox is empty |
 | 15 | n8n Self-Health Monitor | Scheduled (every 15 min) | Checks n8n execution history for failures in the past hour; deduplicates via Postgres; pushes Gotify alert with workflow name and error summary |
 | 16 | NFS Mount Watchdog | Scheduled (every 5 min) | Probes NFS mount points on PD; if missing: runs mount script, re-probes, restarts affected containers (Plex, Audiobookshelf, Immich); notifies via Gotify on heal or escalation |
@@ -188,7 +188,7 @@ curl -X POST https://n8n.paccoco.com/webhook/transcribe/summarize \

 ### Add a class recording (just drop a file):
 ```
-/mnt/data/class-recordings/
+/mnt/tank/class-recordings/
 ├── CS101-Java/
 │   ├── lecture-01-intro.wav
 │   ├── lecture-02-variables.wav
@@ -199,8 +199,9 @@ curl -X POST https://n8n.paccoco.com/webhook/transcribe/summarize \
    └── essay-workshop.wav
 ```

-The workflow detects the class from the subfolder name and the lecture title from the filename.
-Just drop a `.wav` file into the right class folder and it auto-processes.
+Workflow 08 uses a webhook trigger (`POST /webhook/class/upload`), not a folder watcher. Upload the file via curl and include the class name and lecture title as form fields (see curl example below).
+
+> **Note:** The folder structure above shows suggested organization on disk. The workflow does not watch the folder — you must POST to the webhook.

 ### Query RAG for class-specific content:
 ```bash
@@ -213,20 +214,31 @@ The RAG system stores class name metadata with each chunk, so when you mention a

 ### Class Recordings Setup (Workflow 08)

-Requires n8n volume mount to the NFS share. Add to n8n's docker-compose:
-```yaml
-volumes:
-  - /mnt/tank/docker/appdata/n8n:/home/node/.n8n
-  - /mnt/data/class-recordings:/data/class-recordings
-```
-
-Create class subfolders:
+Create class subfolders on PD (already done for CS101-Java):
 ```bash
-mkdir -p /mnt/data/class-recordings/CS101-Java
-mkdir -p /mnt/data/class-recordings/MATH201
+sudo mkdir -p /mnt/tank/class-recordings/CS101-Java
+sudo mkdir -p /mnt/tank/class-recordings/MATH201
 ```

-Then restart n8n and activate the workflow. Any `.wav` dropped into a subfolder triggers automatic transcription + RAG ingest.
+Add to n8n's **automation/.env**:
+```
+PAPERLESS_TAG_LECTURE_NOTES=11
+```
+
+Add to n8n's **automation/docker-compose.yaml** under the n8n `environment:` section:
+```yaml
+      PAPERLESS_TAG_LECTURE_NOTES: ${PAPERLESS_TAG_LECTURE_NOTES}
+```
+
+Then restart n8n and import/activate `08-class-recording-rag-v1.1.json`.
+
+To upload a class recording:
+```bash
+curl -X POST https://n8n.paccoco.com/webhook/class/upload \
+  -F "data=@lecture-01-intro.wav" \
+  -F "class_name=CS101-Java" \
+  -F "lecture_title=lecture-01-intro"
+```

 ### Trigger RSS digest manually:
 ```bash