Arcmira: YouTube Transcript Search to CSV
Arcmira https://arcmira.com
API docs https://arcmira.com/docs
Source and setup https://github.com/arcmira/integrations/tree/master/examples/node-red-source-shortlist
Search indexed YouTube videos for spoken passages and export a CSV source shortlist with timestamps, source links and a JSON file containing coverage notes. Uses built-in Node-RED nodes and your Arcmira account.
Set ARCMIRA_API_KEY privately in the runtime environment and ARCMIRA_OUTPUT_DIR to an existing writable local directory. Import the flow, edit topic and date range in Search input, deploy, and trigger once. Each run makes one search with limit=3 and overwrites the two output files. See the setup guide for current usage limits and file-error behavior.
The flow passed native Node-RED5.0.7 tests and one Free-account search. Coverage is selective; timestamps require playback review before use as edit boundaries. Apache-2.0 for the flow; third-party media rights remain separate.
[{"id":"arcmira-tab","type":"tab","label":"Arcmira: YouTube Transcript Search","disabled":false,"info":"One manual bounded search; CSV plus JSON coverage sidecar. Set private process environment ARCMIRA_API_KEY and ARCMIRA_OUTPUT_DIR. See README.","env":[]},{"id":"arcmira-guide","type":"comment","z":"arcmira-tab","name":"Set private environment, edit dates, then click once","info":"Homepage: https://arcmira.com\nAPI docs: https://arcmira.com/docs\nSet ARCMIRA_API_KEY privately in the runtime environment, not this flow. Set an existing ARCMIRA_OUTPUT_DIR. Edit Search input topic/after/before. Files with the same name are overwritten on each run. JSON keeps coverage even with empty results. Errors clear CSV candidates. No scheduling/retries/Premium/writes/downloads.","x":310,"y":60,"wires":[]},{"id":"arcmira-input","type":"inject","z":"arcmira-tab","name":"Search input (manual)","props":[{"p":"payload"}],"repeat":"","crontab":"","once":false,"onceDelay":0.1,"topic":"","payload":"{\"topic\": \"creator economy\", \"after\": \"2026-08-01\", \"before\": \"2026-09-01\"}","payloadType":"json","x":160,"y":140,"wires":[["arcmira-request"]]},{"id":"arcmira-request","type":"function","z":"arcmira-tab","name":"Validate and prepare one GET","func":"const input = msg.payload || {};\nconst requested = { q: typeof input.topic === \"string\" ? input.topic.trim() : \"\", after: input.after, before: input.before, limit: 3 };\nconst outputDir = env.get(\"ARCMIRA_OUTPUT_DIR\");\nfunction invalid(code) {\n return [null, { requested, outputDir, failure: { code, message: \"Check the input and local environment; no request was sent.\" } }];\n}\nif (typeof outputDir !== \"string\" || !outputDir.trim()) return invalid(\"missing_output_directory\");\nconst key = env.get(\"ARCMIRA_API_KEY\");\nif (typeof key !== \"string\" || !key.trim()) return invalid(\"missing_api_key\");\nfunction validDate(value) {\n if (typeof value !== \"string\" || !/^\\d{4}-\\d{2}-\\d{2}$/.test(value)) return false;\n const date = new Date(value + \"T00:00:00Z\");\n return Number.isFinite(date.getTime()) && date.toISOString().slice(0, 10) === value;\n}\nif (requested.q.length < 2 || !validDate(requested.after) || !validDate(requested.before) || requested.after >= requested.before) return invalid(\"invalid_search_input\");\nconst query = Object.entries(requested).map(([name, value]) => encodeURIComponent(name) + \"=\" + encodeURIComponent(value)).join(\"&\");\nreturn [{\n requested, outputDir,\n url: \"https://api.arcmira.com/v1/search?\" + query,\n headers: { Authorization: \"Bearer \" + key.trim(), Accept: \"application/json\" },\n followRedirects: false,\n requestTimeout: 15000\n}, null];","outputs":2,"timeout":0,"noerr":0,"initialize":"","finalize":"","libs":[],"x":420,"y":140,"wires":[["arcmira-http"],["arcmira-build"]]},{"id":"arcmira-http","type":"http request","z":"arcmira-tab","name":"Search Arcmira (limit 3)","method":"GET","ret":"obj","paytoqs":"ignore","url":"","tls":"","persist":false,"proxy":"","insecureHTTPParser":false,"authType":"","senderr":true,"headers":[],"x":700,"y":140,"wires":[["arcmira-build"]]},{"id":"arcmira-catch","type":"catch","z":"arcmira-tab","name":"Handle transport failure","scope":["arcmira-http"],"uncaught":false,"x":690,"y":220,"wires":[["arcmira-build"]]},{"id":"arcmira-build","type":"function","z":"arcmira-tab","name":"Build safe CSV and source packet","func":"const requested = msg.requested || null;\nconst response = msg.payload;\nlet failure = msg.failure || null;\nconst statusCode = Number(msg.statusCode);\nif (!failure && (msg.error || !Number.isFinite(statusCode))) {\n failure = { code: \"transport_error\", message: \"The request did not complete. No retry was attempted.\" };\n}\nif (!failure && statusCode !== 200) {\n const code = response && response.error && response.error.code;\n failure = { code: typeof code === \"string\" && /^[a-z_]{1,80}$/.test(code) ? code : \"http_error\", http_status: statusCode, message: \"Arcmira refused the request. Check API error documentation; no retry or query change was attempted.\" };\n}\nif (!failure && (!response || !Array.isArray(response.chunks) || response.chunks.length > 3)) {\n failure = { code: \"invalid_response\", message: \"Expected at most three passages; no source sheet was produced.\" };\n}\nif (!failure && response.chunks.some(c => !c || typeof c.video_id !== \"string\" || typeof c.text !== \"string\" || typeof c.watch_url !== \"string\" || (c.start_seconds !== null && (!Number.isFinite(c.start_seconds) || c.start_seconds < 0)))) {\n failure = { code: \"invalid_passage\", message: \"A passage lacks required source fields; no source sheet was produced.\" };\n}\nconst packet = {\n status: failure ? \"error\" : response.chunks.length ? \"ok\" : \"empty\",\n query: requested ? requested.q : null,\n requested_window: requested ? { after: requested.after, before: requested.before } : null,\n applied_window: failure ? null : response.window ?? null,\n partial: failure ? null : response.partial ?? null,\n failed_batches: failure ? null : response.failed_batches ?? null,\n as_of: failure ? null : response.as_of ?? null,\n search_index: failure ? null : response.search_index ?? null,\n note: failure ? null : response.note ?? null,\n filters: failure ? null : response.filters ?? null,\n access: failure ? null : response.access ?? null,\n returned: failure ? 0 : response.chunks.length,\n error: failure,\n candidates: failure ? [] : response.chunks.map((c, i) => ({\n ordinal: i + 1, video_id: c.video_id, title: c.video_title ?? null,\n channel: c.channel_name ?? null, published_at: c.published_at ?? null,\n passage: c.text, passage_start_seconds: c.start_seconds,\n source_watch_url: c.watch_url,\n arcmira_url: c.watch_url.startsWith(\"/\") && !c.watch_url.startsWith(\"//\") ? \"https://arcmira.com\" + c.watch_url : c.watch_url,\n original_source_url: \"https://www.youtube.com/watch?v=\" + encodeURIComponent(c.video_id) + (c.start_seconds === null ? \"\" : \"&t=\" + c.start_seconds + \"s\"),\n transcript_source: c.source ?? null, transcript_source_label: c.source_label ?? null,\n speaker_names: c.speakers ?? [], speaker_entities: c.speakers_by ?? [],\n cut_start_seconds: null, cut_end_seconds: null, playback_verified: false,\n speaker_identity_verified: false, reuse_permission: \"not established\"\n }))\n};\nconst columns = [\"ordinal\", \"video_id\", \"title\", \"channel\", \"published_at\", \"passage\", \"passage_start_seconds\", \"source_watch_url\", \"arcmira_url\", \"original_source_url\", \"transcript_source\", \"transcript_source_label\", \"speaker_names\", \"speaker_entities\", \"requested_window\", \"applied_window\", \"partial\", \"search_index\", \"note\", \"cut_start_seconds\", \"cut_end_seconds\", \"playback_verified\", \"speaker_identity_verified\", \"reuse_permission\"];\nfunction cell(value) {\n let text = value == null ? \"\" : typeof value === \"object\" ? JSON.stringify(value) : String(value);\n // Quote every cell; prefix spreadsheet formula starters, including hidden whitespace.\n if (/^[\\s\\u0000-\\u001f\\u007f]*[=+@-]/u.test(text)) text = \"'\" + text;\n return '\"' + text.replace(/\"/g, '\"\"') + '\"';\n}\nconst csv = [columns.map(cell).join(\",\"), ...packet.candidates.map(candidate => columns.map(name => cell(Object.prototype.hasOwnProperty.call(candidate, name) ? candidate[name] : packet[name])).join(\",\"))].join(\"\\r\\n\") + \"\\r\\n\";\nconst summary = { payload: { status: packet.status, returned: packet.returned, error: failure, instruction: \"Use the JSON sidecar for date window and coverage. Verify sources and rights before editing.\" } };\nif (typeof msg.outputDir !== \"string\" || !msg.outputDir.trim()) return [null, null, summary];\nconst prefix = msg.outputDir.replace(/[\\\\/]$/, \"\") + \"/arcmira-source-shortlist\";\nreturn [\n { payload: csv, filename: prefix + \".csv\" },\n { payload: JSON.stringify(packet, null, 2) + \"\\n\", filename: prefix + \".json\" },\n summary\n];","outputs":3,"timeout":0,"noerr":0,"initialize":"","finalize":"","libs":[],"x":1000,"y":140,"wires":[["arcmira-csv"],["arcmira-json"],["arcmira-status"]]},{"id":"arcmira-csv","type":"file","z":"arcmira-tab","name":"Save CSV (overwrite)","filename":"filename","filenameType":"msg","appendNewline":false,"createDir":false,"overwriteFile":"true","encoding":"utf8","x":1320,"y":100,"wires":[[]]},{"id":"arcmira-json","type":"file","z":"arcmira-tab","name":"Save JSON sidecar (overwrite)","filename":"filename","filenameType":"msg","appendNewline":false,"createDir":false,"overwriteFile":"true","encoding":"utf8","x":1350,"y":160,"wires":[[]]},{"id":"arcmira-status","type":"debug","z":"arcmira-tab","name":"Result summary (not raw request)","active":true,"tosidebar":true,"console":false,"tostatus":false,"complete":"payload","targetType":"msg","statusVal":"","statusType":"auto","x":1370,"y":240,"wires":[]},{"id":"arcmira-write-catch","type":"catch","z":"arcmira-tab","name":"Report file or flow failure","scope":["arcmira-request","arcmira-build","arcmira-csv","arcmira-json"],"uncaught":false,"x":990,"y":320,"wires":[["arcmira-safe-error"]]},{"id":"arcmira-safe-error","type":"function","z":"arcmira-tab","name":"Safe local error","func":"return { payload: { status: \"error\", code: \"local_flow_or_file_error\", message: \"No retry was attempted. Check your local output directory and flow configuration; discard incomplete output files.\" } };","outputs":1,"timeout":0,"noerr":0,"initialize":"","finalize":"","libs":[],"x":1270,"y":320,"wires":[["arcmira-status"]]}]
