{ "name": "Turn YouTube and Bilibili videos into a blog post, X thread and LinkedIn post with Claude via apimodels.app", "nodes": [ { "parameters": { "content": "## Turn a video into a blog post, X thread and LinkedIn post\n\nPaste a YouTube or Bilibili link and get a ready-to-edit blog post, an X thread, a LinkedIn post, a newsletter blurb and an optional 16:9 cover image. The subtitles, the writing (Claude Sonnet 5.5) and the cover (GPT Image 2) all come from one API key on [apimodels.app](https://apimodels.app), so there is no separate transcript service to sign up for.\n\n### How it works\n1. The form takes the video link, output language, tone and whether to make a cover.\n2. **Get subtitles** lists the video's caption tracks; **Pick subtitle track** prefers human captions in the spoken language.\n3. The captions are downloaded and cleaned into plain text.\n4. Claude writes every format in one call and returns JSON.\n5. If requested, GPT Image 2 draws a cover with the headline on it.\n6. The result page shows everything in copy-ready boxes plus the real cost of the run.\n\n### Setup\n1. Create an API key in the [apimodels console](https://apimodels.app/console/api-keys). New accounts get $0.1 of free credit.\n2. Create a **Bearer Auth** credential with that key and select it in the three apimodels HTTP Request nodes.\n3. Activate the workflow and open the production form URL from the trigger node.\nStep-by-step guide: [apimodels.app/docs/n8n](https://apimodels.app/docs/n8n)\n\n### Customize\n- Change the writing model or transcript limit in **Settings**.\n- Edit lengths and formats in the system message of **Write posts**.\n- Only videos with captions work (most YouTube videos have auto-generated ones).", "height": 900, "width": 540, "color": 1 }, "id": "083e3656-39c6-56c2-9f9b-d5a99a0e1b14", "name": "Overview", "type": "n8n-nodes-base.stickyNote", "typeVersion": 1, "position": [ -660, -320 ] }, { "parameters": { "content": "## 1. Get the transcript\nLists caption tracks, picks the best one, downloads and cleans it.", "height": 420, "width": 1120, "color": 7 }, "id": "121a415c-290b-57fa-a0d6-96fde927a481", "name": "Section: subtitles", "type": "n8n-nodes-base.stickyNote", "typeVersion": 1, "position": [ -40, 20 ] }, { "parameters": { "content": "## 2. Write and illustrate\nOne Claude call writes all formats as JSON; GPT Image 2 makes the cover if asked.", "height": 420, "width": 1240, "color": 7 }, "id": "babf539a-3365-5afb-a2b5-f4018344dd89", "name": "Section: write", "type": "n8n-nodes-base.stickyNote", "typeVersion": 1, "position": [ 1120, 20 ] }, { "parameters": { "content": "## 3. Result pages\nCopy-ready text with the run's cost, or the provider's own error message.", "height": 760, "width": 480, "color": 7 }, "id": "02ab449b-e18c-574b-b7e1-107ac7a30e93", "name": "Section: result", "type": "n8n-nodes-base.stickyNote", "typeVersion": 1, "position": [ 2400, -180 ] }, { "parameters": { "formTitle": "Video to posts", "formDescription": "Paste a YouTube or Bilibili link. You get a blog post, an X thread, a LinkedIn post and a newsletter blurb written from the video's subtitles.", "formFields": { "values": [ { "fieldLabel": "Video URL", "requiredField": true, "placeholder": "https://www.youtube.com/watch?v=..." }, { "fieldLabel": "Output language", "fieldType": "dropdown", "requiredField": true, "fieldOptions": { "values": [ { "option": "Same as the video" }, { "option": "English" }, { "option": "Chinese (Simplified)" }, { "option": "Spanish" }, { "option": "Portuguese" }, { "option": "German" }, { "option": "French" }, { "option": "Japanese" } ] } }, { "fieldLabel": "Tone", "fieldType": "dropdown", "requiredField": true, "fieldOptions": { "values": [ { "option": "Professional" }, { "option": "Friendly" }, { "option": "Witty" } ] } }, { "fieldLabel": "Cover image", "fieldType": "dropdown", "requiredField": true, "fieldOptions": { "values": [ { "option": "Yes, make a 16:9 cover" }, { "option": "No cover" } ] } }, { "fieldLabel": "Notes for the writer", "fieldType": "textarea", "placeholder": "Optional: audience, a link to include, words to avoid" } ] }, "options": { "path": "video-to-posts", "buttonLabel": "Write the posts" } }, "id": "f5c9df2d-0c37-5715-9b4c-2022f67e073e", "name": "Video link form", "type": "n8n-nodes-base.formTrigger", "typeVersion": 2.2, "position": [ 0, 240 ], "webhookId": "8ae58c6a-44f9-5418-8b27-fef2730f7937" }, { "parameters": { "assignments": { "assignments": [ { "id": "c9fa8713-6b35-523c-8e70-64803710a0d5", "name": "api_base", "value": "https://api.apimodels.app/v1", "type": "string" }, { "id": "21e0b2aa-f67a-5f4b-99bc-deaa64572f71", "name": "writer_model", "value": "claude-sonnet-5-5", "type": "string" }, { "id": "024511e3-2f13-57e4-91c4-49905bd078c9", "name": "image_model", "value": "gpt-image-2", "type": "string" }, { "id": "f4b11d21-5cb9-52a0-8308-16b72fbdd9d3", "name": "max_transcript_chars", "value": 60000, "type": "number" } ] }, "includeOtherFields": true, "options": {} }, "id": "a10e316f-966f-5337-bdd1-3bffce8a5edf", "name": "Settings", "type": "n8n-nodes-base.set", "typeVersion": 3.4, "position": [ 220, 240 ] }, { "parameters": { "method": "POST", "url": "={{ $json.api_base }}/media/subtitles", "authentication": "genericCredentialType", "genericAuthType": "httpBearerAuth", "sendBody": true, "specifyBody": "json", "jsonBody": "={{ JSON.stringify({ url: $json['Video URL'].trim() }) }}", "options": { "response": { "response": { "fullResponse": true } }, "timeout": 150000 } }, "id": "fb7a97d0-cec8-5b3b-8c47-82df6d051b04", "name": "Get subtitles", "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.2, "position": [ 440, 240 ], "credentials": { "httpBearerAuth": { "id": "REPLACE_WITH_YOUR_CREDENTIAL", "name": "apimodels API key" } }, "onError": "continueErrorOutput" }, { "parameters": { "jsCode": "// Prefer human captions in the spoken language. The auto-generated track tells us which language is spoken.\nconst res = $input.first().json;\nconst data = (res.body && res.body.data) || {};\nconst tracks = data.subtitles || [];\nif (!tracks.length) {\n throw new Error('This video has no subtitles. Try a YouTube or Bilibili video with captions.');\n}\nconst isAuto = (t) => /auto-generated/i.test(t.language_name || '');\nconst auto = tracks.find(isAuto);\nconst manual = tracks.filter((t) => !isAuto(t));\nconst pick = manual.find((t) => auto && t.language_tag === auto.language_tag) || auto || manual.find((t) => /^en/.test(t.language_tag)) || manual[0];\nconst order = ['srt', 'vtt', 'json', 'json3'];\nconst file = order.map((f) => (pick.urls || []).find((u) => u.format === f)).find(Boolean) || pick.urls[0];\n\nreturn [{ json: {\n title: data.title || '',\n language: pick.language_name,\n format: file.format,\n subtitle_url: file.url,\n subtitles_cost: Number(res.headers['x-apimodels-cost'] || 0),\n} }];" }, "id": "afb1241a-8384-5277-b872-dd5726353b99", "name": "Pick subtitle track", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ 660, 240 ], "onError": "continueErrorOutput" }, { "parameters": { "url": "={{ $json.subtitle_url }}", "options": { "response": { "response": { "responseFormat": "text" } }, "timeout": 60000 } }, "id": "4e06a434-1dae-57ab-9d4d-d4be5efab34a", "name": "Download subtitles", "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.2, "position": [ 880, 240 ], "onError": "continueErrorOutput" }, { "parameters": { "jsCode": "// SRT / VTT -> plain text; Bilibili returns JSON with body[].content.\nconst meta = $('Pick subtitle track').first().json;\nconst raw = String($input.first().json.data || '');\nlet lines = [];\nif (raw.trim().startsWith('{')) {\n try {\n const j = JSON.parse(raw);\n lines = (j.body || []).map((x) => x.content).filter(Boolean);\n if (!lines.length && j.events) lines = j.events.map((e) => (e.segs || []).map((s) => s.utf8).join('')).filter(Boolean);\n } catch (e) { /* fall through to text parsing */ }\n}\nif (!lines.length) {\n lines = raw.split(/\\r?\\n/)\n .map((l) => l.replace(/<[^>]+>/g, '').trim())\n .filter((l) => l && !/^\\d+$/.test(l) && !l.includes('-->') && !/^(WEBVTT|Kind:|Language:)/.test(l));\n}\n// Auto captions repeat the same line; keep each line once in a row.\nconst kept = [];\nfor (const l of lines) if (l !== kept[kept.length - 1]) kept.push(l);\nconst max = $('Settings').first().json.max_transcript_chars;\nconst transcript = kept.join(' ').replace(/\\s+/g, ' ').trim().slice(0, max);\nif (transcript.length < 200) throw new Error('The subtitles are too short to write from.');\n\nreturn [{ json: { transcript, chars: transcript.length, title: meta.title } }];" }, "id": "3a09dc58-89af-54d4-a48b-515446bb7dfb", "name": "Clean transcript", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ 1100, 240 ], "onError": "continueErrorOutput" }, { "parameters": { "method": "POST", "url": "={{ $('Settings').first().json.api_base }}/chat/completions", "authentication": "genericCredentialType", "genericAuthType": "httpBearerAuth", "sendBody": true, "specifyBody": "json", "jsonBody": "={{ (() => {\n const form = $('Video link form').first().json;\n const lang = form['Output language'] === 'Same as the video' ? 'the same language as the transcript' : form['Output language'];\n const system = [\n 'You repurpose video transcripts into written content. Reply with one JSON object and nothing else, with these keys:',\n 'blog_title (string);',\n 'blog_markdown (string, 600 to 900 words, use ## section headings, no # title, no links you were not given);',\n 'x_thread (array of 5 to 8 strings, each under 270 characters, the first one is a hook, at most one hashtag in the last one);',\n 'linkedin_post (string, 120 to 220 words, short paragraphs);',\n 'newsletter_blurb (string, 60 to 100 words);',\n 'cover_prompt (string, an English prompt for a bold, simple 16:9 blog cover about the topic that renders a headline of at most 6 words in large clean letters; give the exact headline in quotes).',\n 'Stay faithful to the transcript: do not invent facts, numbers, names or quotes.',\n `Write everything except cover_prompt in ${lang}. Tone: ${form['Tone']}.`\n ].join(' ');\n const user = `Video title: ${$json.title}\\nNotes from the user: ${form['Notes for the writer'] || 'none'}\\n\\nTranscript:\\n${$json.transcript}`;\n return JSON.stringify({ model: $('Settings').first().json.writer_model, max_tokens: 6000, messages: [{ role: 'system', content: system }, { role: 'user', content: user }] });\n})() }}", "options": { "response": { "response": { "fullResponse": true } }, "timeout": 240000 } }, "id": "704a1f23-fb12-57c8-a35f-b5586753708a", "name": "Write posts", "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.2, "position": [ 1340, 240 ], "credentials": { "httpBearerAuth": { "id": "REPLACE_WITH_YOUR_CREDENTIAL", "name": "apimodels API key" } }, "onError": "continueErrorOutput" }, { "parameters": { "jsCode": "const res = $input.first().json;\nconst text = (res.body && res.body.choices && res.body.choices[0].message.content) || '';\nconst match = text.match(/\\{[\\s\\S]*\\}/);\nlet posts;\ntry {\n posts = JSON.parse(match ? match[0] : text);\n} catch (e) {\n throw new Error('The writing model did not return valid JSON. Run it again or pick another model in Settings.');\n}\nreturn [{ json: { ...posts, writer_cost: Number(res.headers['x-apimodels-cost'] || 0) } }];" }, "id": "3591a5b2-3b6a-5472-8efb-c89d9748c706", "name": "Parse posts", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ 1560, 240 ], "onError": "continueErrorOutput" }, { "parameters": { "conditions": { "options": { "caseSensitive": true, "leftValue": "", "typeValidation": "strict", "version": 2 }, "conditions": [ { "id": "0c70979d-4c8d-56d8-a4a5-89a8876fb8a6", "leftValue": "={{ $('Video link form').first().json['Cover image'].startsWith('Yes') }}", "operator": { "type": "boolean", "operation": "true", "singleValue": true }, "rightValue": "" } ], "combinator": "and" }, "options": {} }, "id": "afe37c04-37f8-50f0-b422-7e26cd1f4da5", "name": "Make a cover?", "type": "n8n-nodes-base.if", "typeVersion": 2.2, "position": [ 1780, 240 ] }, { "parameters": { "method": "POST", "url": "={{ $('Settings').first().json.api_base }}/images/generations", "authentication": "genericCredentialType", "genericAuthType": "httpBearerAuth", "sendBody": true, "specifyBody": "json", "jsonBody": "={{ JSON.stringify({ model: $('Settings').first().json.image_model, prompt: $json.cover_prompt, size: '1536x1024', response_format: 'url' }) }}", "options": { "response": { "response": { "fullResponse": true } }, "timeout": 300000 } }, "id": "53c786e3-28a1-5f32-b228-bfc1411391cf", "name": "Generate cover", "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.2, "position": [ 2000, 120 ], "credentials": { "httpBearerAuth": { "id": "REPLACE_WITH_YOUR_CREDENTIAL", "name": "apimodels API key" } }, "onError": "continueRegularOutput" }, { "parameters": { "url": "={{ $('Settings').first().json.api_base }}/records/{{ (($json.body?.data?.[0]?.url || '').split('/').pop() || 'none').split('.')[0] }}", "authentication": "genericCredentialType", "genericAuthType": "httpBearerAuth", "options": { "timeout": 30000 } }, "id": "40f4d704-09c0-5002-a666-6851b1ce6533", "name": "Get cover cost", "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.2, "position": [ 2220, 120 ], "credentials": { "httpBearerAuth": { "id": "REPLACE_WITH_YOUR_CREDENTIAL", "name": "apimodels API key" } }, "onError": "continueRegularOutput" }, { "parameters": { "jsCode": "const esc = (s) => String(s ?? '').replace(/&/g, '&').replace(//g, '>').replace(/\"/g, '"');\nconst posts = $('Parse posts').first().json;\nconst sub = $('Pick subtitle track').first().json;\nconst blog = posts.blog_markdown || '';\n// CJK text has no spaces between words, so count characters there.\nconst cjk = (blog.match(/[\\u3040-\\u30ff\\u3400-\\u9fff\\uac00-\\ud7af]/g) || []).length;\nconst length = cjk > blog.length / 4 ? `${cjk} characters` : `${blog.split(/\\s+/).filter(Boolean).length} words`;\n\nlet coverHtml = '';\nlet coverCost = 0;\nif ($('Generate cover').isExecuted) {\n const r = $('Generate cover').first().json;\n const url = r.body && r.body.data && r.body.data[0] && r.body.data[0].url;\n const rec = $('Get cover cost').isExecuted ? $('Get cover cost').first().json : {};\n coverCost = Number((rec.data && rec.data.credits) || 0);\n coverHtml = url\n ? `

Cover

\"Cover

Open full size

`\n : `

Cover

The cover could not be made: ${esc((r.error && r.error.message) || 'no image returned')}

`;\n}\n\nconst box = (label, text, rows) => `

${label}

`;\nconst thread = (posts.x_thread || []).map((t, i) => `
  • ${i + 1}/${posts.x_thread.length} · ${String(t).length} chars
    ${esc(t)}
    • `).join('');\nconst total = sub.subtitles_cost + posts.writer_cost + coverCost;\n\nconst html = `\n\n
    \n

    ${esc(posts.blog_title || sub.title)}

    \n

    From \"${esc(sub.title)}\" · ${esc(sub.language)} subtitles · ${length} · charged $${total.toFixed(4)} (subtitles $${sub.subtitles_cost}, writing $${posts.writer_cost}${coverCost ? `, cover $${coverCost}` : ''})

    \n${coverHtml}\n${box('Blog post (Markdown)', posts.blog_markdown, 18)}\n

    X thread

      ${thread}
    \n${box('LinkedIn post', posts.linkedin_post, 9)}\n${box('Newsletter blurb', posts.newsletter_blurb, 5)}\n

    Do another video

    \n
    `;\n\nreturn [{ json: { html } }];" }, "id": "88906dfa-c8ad-5587-a760-569e21b4b249", "name": "Build result page", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ 2440, 240 ] }, { "parameters": { "operation": "completion", "respondWith": "showText", "responseText": "={{ $json.html }}" }, "id": "73cd0224-c41f-5c30-9525-5e153a26d467", "name": "Show result", "type": "n8n-nodes-base.form", "typeVersion": 1, "position": [ 2660, 240 ], "webhookId": "73cd0224-c41f-5c30-9525-5e153a26d467" }, { "parameters": { "operation": "completion", "respondWith": "showText", "responseText": "=\n
    \n

    The posts could not be made

    \n

    {{ $json.error?.message || $json.error || 'Something went wrong.' }}

    \n

    Only videos with subtitles work. Failed requests are not charged.

    \nTry another video\n
    " }, "id": "4eaf5824-47ce-5c2b-b52f-85182c1a6cad", "name": "Show error", "type": "n8n-nodes-base.form", "typeVersion": 1, "position": [ 2660, 460 ], "webhookId": "4eaf5824-47ce-5c2b-b52f-85182c1a6cad" } ], "connections": { "Video link form": { "main": [ [ { "node": "Settings", "type": "main", "index": 0 } ] ] }, "Settings": { "main": [ [ { "node": "Get subtitles", "type": "main", "index": 0 } ] ] }, "Get subtitles": { "main": [ [ { "node": "Pick subtitle track", "type": "main", "index": 0 } ], [ { "node": "Show error", "type": "main", "index": 0 } ] ] }, "Pick subtitle track": { "main": [ [ { "node": "Download subtitles", "type": "main", "index": 0 } ], [ { "node": "Show error", "type": "main", "index": 0 } ] ] }, "Download subtitles": { "main": [ [ { "node": "Clean transcript", "type": "main", "index": 0 } ], [ { "node": "Show error", "type": "main", "index": 0 } ] ] }, "Clean transcript": { "main": [ [ { "node": "Write posts", "type": "main", "index": 0 } ], [ { "node": "Show error", "type": "main", "index": 0 } ] ] }, "Write posts": { "main": [ [ { "node": "Parse posts", "type": "main", "index": 0 } ], [ { "node": "Show error", "type": "main", "index": 0 } ] ] }, "Parse posts": { "main": [ [ { "node": "Make a cover?", "type": "main", "index": 0 } ], [ { "node": "Show error", "type": "main", "index": 0 } ] ] }, "Make a cover?": { "main": [ [ { "node": "Generate cover", "type": "main", "index": 0 } ], [ { "node": "Build result page", "type": "main", "index": 0 } ] ] }, "Generate cover": { "main": [ [ { "node": "Get cover cost", "type": "main", "index": 0 } ] ] }, "Get cover cost": { "main": [ [ { "node": "Build result page", "type": "main", "index": 0 } ] ] }, "Build result page": { "main": [ [ { "node": "Show result", "type": "main", "index": 0 } ] ] } }, "pinData": {}, "settings": { "executionOrder": "v1" }, "meta": { "templateCredsSetupCompleted": false }, "tags": [] }