Update to HappyScribe
Snapshot Oct 28, 2026 · 12:08 UTC · version 1.1.0
Package or technical metadata updated
Package or technical metadata updated
Observed in package metadata. These changes alone do not establish a new customer-facing feature.
Up to 3 plugins free · no card required.
All fields and original values.
These are saved file metadata. Open a changed file to read its text edits.
Package file
Sep 30, 2026 · 22:47 UTC
62fea515a5b51acb2f5cd294fecdc62890754175f00c115ced0851de3b9d35fa
Oct 28, 2026 · 12:08 UTC
3bd8fab65d4855054391e5f42091c43f14eb0259f7f2a0ad86a0ac051dc0dd7e
Package file
Sep 30, 2026 · 22:47 UTC
127
Oct 28, 2026 · 12:08 UTC
103
Package file
Sep 30, 2026 · 22:47 UTC
e0016fe04e230a5561a6d4ec2cf8d62476873377f44eeeb674b9829a93cbcf33
Oct 28, 2026 · 12:08 UTC
0052e36bdacfdaeb306a9f1ae6108a47eee24a4285cd698ae943ae18b64c0786
Package file
Sep 30, 2026 · 22:47 UTC
2915
Oct 28, 2026 · 12:08 UTC
7018
Compare saved observations
Download comparison JSONChanged files
Choose a file to see its text changes. Binary or large files show metadata only.
assets/screenshot-1-transcribe.png →
assets/screenshot-2-transcript.png →
assets/screenshot-3-meetings.png →
.codex-plugin/plugin.json
− removed · + added
--- before +++ after @@ -1,26 +1,101 @@ { - "apps": "./.app.json", "author": { "name": "HappyScribe" }, - "description": "Connect HappyScribe to give ChatGPT on-demand access to your organization's conversational memory: every transcript, meeting recording, interview, and quote, speaker-attributed, verbatim, in 150+ languages.\n\nHappyScribe is an AI notetaker and transcription platform that captures every conversation: online meetings, in-person recordings, interviews, fieldwork, uploaded media. It turns them into searchable, citable, structured knowledge with verbatim accuracy: what was actually said, by whom, in what language.\n\nWhat ChatGPT can do with HappyScribe connected:\n- Search your library by keyword, speaker, date, or topic\n- Pull exact quotes with verified timestamps and deep links to the editor\n- Transcribe new media from a URL (YouTube, Vimeo, direct links)\n- Manage folders, rename speakers, and edit transcript text\n- Diagnose why a meeting wasn't recorded\n- Explore the people and companies graph across your recordings\n\nBest-in-class multilingual accuracy across 150+ languages. European and GDPR-native.", + "description": "Bring your HappyScribe transcripts and meetings into ChatGPT. Attach a recording, paste a link or drop files into the HappyScribe upload box right in the chat, follow the transcription, and open the finished transcript in a card with its audio player.\n\nHappyScribe is an AI notetaker and transcription platform. It captures online meetings, in-person recordings, interviews and uploaded media, and turns them into accurate, speaker-labeled transcripts in 150+ languages.\n\nWhat ChatGPT can do with HappyScribe connected:\n- Transcribe recordings you attach, link or upload, and track their progress\n- Search your whole library by keyword, speaker, date or topic, including meetings recorded by the HappyScribe notetaker\n- Pull exact quotes with verified timestamps and links to the moment in the recording\n- See what was on screen in a video recording: slides, demos and screen shares\n- Fix transcripts by asking: rename speakers, correct a name everywhere at once, keep terms right with glossaries\n- Export to Word, PDF, text, spreadsheets or subtitles (SRT, VTT), and share transcripts with your team\n\nHappyScribe only accesses the workspace you connect. European and GDPR-native.", "interface": { - "capabilities": [], + "capabilities": [ + "Transcribe audio and video", + "Search transcripts and meetings", + "Verified quotes with timestamps", + "Export and share transcripts" + ], "category": "Productivity", "defaultPrompt": [ - "hey chat, what did the design team discuss in their last meeting?", - "what has been the team up to since I went on vacation on December 21st?", - "help me understand what this file in afrikaans is about" + "Transcribe this recording and show me the transcript", + "What did we decide in this week's meetings?", + "Pull the three best quotes from my last interview, with timestamps" ], "developerName": "HappyScribe", "displayName": "HappyScribe", - "longDescription": "Connect HappyScribe to give ChatGPT on-demand access to your organization's conversational memory: every transcript, meeting recording, interview, and quote, speaker-attributed, verbatim, in 150+ languages.\n\nHappyScribe is an AI notetaker and transcription platform that captures every conversation: online meetings, in-person recordings, interviews, fieldwork, uploaded media. It turns them into searchable, citable, structured knowledge with verbatim accuracy: what was actually said, by whom, in what language.\n\nWhat ChatGPT can do with HappyScribe connected:\n- Search your library by keyword, speaker, date, or topic\n- Pull exact quotes with verified timestamps and deep links to the editor\n- Transcribe new media from a URL (YouTube, Vimeo, direct links)\n- Manage folders, rename speakers, and edit transcript text\n- Diagnose why a meeting wasn't recorded\n- Explore the people and companies graph across your recordings\n\nBest-in-class multilingual accuracy across 150+ languages. European and GDPR-native.", + "longDescription": "Bring your HappyScribe transcripts and meetings into ChatGPT. Attach a recording, paste a link or drop files into the HappyScribe upload box right in the chat, follow the transcription, and open the finished transcript in a card with its audio player.\n\nHappyScribe is an AI notetaker and transcription platform. It captures online meetings, in-person recordings, interviews and uploaded media, and turns them into accurate, speaker-labeled transcripts in 150+ languages.\n\nWhat ChatGPT can do with HappyScribe connected:\n- Transcribe recordings you attach, link or upload, and track their progress\n- Search your whole library by keyword, speaker, date or topic, including meetings recorded by the HappyScribe notetaker\n- Pull exact quotes with verified timestamps and links to the moment in the recording\n- See what was on screen in a video recording: slides, demos and screen shares\n- Fix transcripts by asking: rename speakers, correct a name everywhere at once, keep terms right with glossaries\n- Export to Word, PDF, text, spreadsheets or subtitles (SRT, VTT), and share transcripts with your team\n\nHappyScribe only accesses the workspace you connect. European and GDPR-native.", "privacyPolicyURL": "https://www.happyscribe.com/privacy", - "shortDescription": "Search transcripts & meetings", + "shortDescription": "Transcribe and search meetings", "supportURL": "https://help.happyscribe.com", "termsOfServiceURL": "https://www.happyscribe.com/terms", - "websiteURL": "https://www.happyscribe.com/" + "websiteURL": "https://www.happyscribe.com/", + "brandColor": "#0B4BD5", + "composerIcon": "./assets/icon.png", + "logo": "./assets/logo.png", + "screenshots": [ + "./assets/screenshot-1-transcribe.png", + "./assets/screenshot-2-transcript.png", + "./assets/screenshot-3-meetings.png", + "./assets/screenshot-4-quotes.png", + "./assets/screenshot-5-edit.png" + ] }, "name": "app-694a8009f86481918b118a4a86bebc8c", - "version": "1.0.0" -} \ No newline at end of file + "version": "1.1.0", + "extensions": { + "com.openai": { + "review": { + "test_cases": { + "positive": [ + { + "description": "Transcribe an attached recording", + "prompt": "Transcribe this recording.", + "tools_triggered": "create_transcription, get_transcription_status", + "expected_behavior": "With an audio file attached, starts one transcription in the test workspace, reports its progress, and confirms when the transcript is ready." + }, + { + "description": "Upload files through the upload box", + "prompt": "I want to upload a recording to transcribe.", + "tools_triggered": "show_upload_box", + "expected_behavior": "Shows the HappyScribe upload box in the chat; files dropped into it are uploaded and start transcribing." + }, + { + "description": "Show a finished transcript", + "prompt": "Show me the transcript of my latest recording.", + "tools_triggered": "list_transcriptions, show_transcript", + "expected_behavior": "Finds the most recent finished transcript in the test workspace and displays it in a card with speakers, timestamps and the audio player." + }, + { + "description": "Pull verified quotes", + "prompt": "Pull three quotes about the roadmap from my latest meeting, with timestamps.", + "tools_triggered": "search_transcriptions, get_transcription, verify_quotes", + "expected_behavior": "Returns quotes that appear word for word in the transcript, each with its timestamp and a link to that moment; nothing is invented." + }, + { + "description": "Export a transcript", + "prompt": "Export my latest transcript as a Word document.", + "tools_triggered": "list_transcriptions, export_transcription, get_export_status", + "expected_behavior": "Creates a Word export of the transcript and returns a download link that expires after 24 hours." + } + ], + "negative": [ + { + "description": "Purchase request", + "prompt": "Buy 10 more hours of transcription minutes for my account." + }, + { + "description": "Unsupported file type", + "prompt": "Transcribe this PDF contract." + }, + { + "description": "Access outside the connected workspace", + "prompt": "Show me the transcripts from another company's workspace." + } + ] + }, + "commerce": false, + "commerce_description": "HappyScribe does not sell anything in ChatGPT. Transcriptions use minutes already included in the user's HappyScribe plan; no purchases or payments happen in the conversation." + }, + "publication": { + "release_notes": "Transcribe files from the chat (attachments, links and an in-chat upload box), live progress, transcript cards with an audio player, exports to Word, PDF, text and subtitles, sharing, glossaries and batch corrections. Updated listing text, starter prompts and screenshots." + } + } + }, + "mcpServers": "./.mcp.json", + "apps": "./.app.json" +}
Full technical diff · 12 changed fields
changed /files/.app.json/sha256
"62fea515a5b51acb2f5cd294fecdc62890754175f00c115ced0851de3b9d35fa"
"3bd8fab65d4855054391e5f42091c43f14eb0259f7f2a0ad86a0ac051dc0dd7e"
changed /files/.app.json/size
127
103
changed /files/.codex-plugin~1plugin.json/sha256
"e0016fe04e230a5561a6d4ec2cf8d62476873377f44eeeb674b9829a93cbcf33"
"0052e36bdacfdaeb306a9f1ae6108a47eee24a4285cd698ae943ae18b64c0786"
changed /files/.codex-plugin~1plugin.json/size
2915
7018
added /files/.mcp.json
Field was absent
{
"sha256": "fb8515d34e2a98b5f96c6fc2833ccfeaa94bf42b6a23b2c1f22c2b1c79f9a5ad",
"size": 133
}added /files/assets~1icon.png
Field was absent
{
"sha256": "00dcd871473684d4e9449009ee90b1b1e0357b2ca6910314e18be09b8c6b6b7c",
"size": 22441
}added /files/assets~1logo.png
Field was absent
{
"sha256": "00dcd871473684d4e9449009ee90b1b1e0357b2ca6910314e18be09b8c6b6b7c",
"size": 22441
}added /files/assets~1screenshot-1-transcribe.png
Field was absent
{
"sha256": "2882cb107877183953e14fbd7d32b5f6e25e448fbff6d77df56d28aed2c81b82",
"size": 95871
}added /files/assets~1screenshot-2-transcript.png
Field was absent
{
"sha256": "a557151d0ffa2270ca276de80b4b8e0d9930dc9daafbfafdd89d92b147609df3",
"size": 105478
}added /files/assets~1screenshot-3-meetings.png
Field was absent
{
"sha256": "e7f66b64a9bade8795174164ba62aa82006c0f3d6d8be28f680422067db4c3d6",
"size": 107022
}added /files/assets~1screenshot-4-quotes.png
Field was absent
{
"sha256": "d7548726886ec777c6e76f8428c730d9f2282ef498699733b1908eba4a457672",
"size": 102265
}added /files/assets~1screenshot-5-edit.png
Field was absent
{
"sha256": "7a66baff1831350136e7eef6768f65f72271ffa9e02d60ee8f97e5e244acbef2",
"size": 90655
}Full snapshot data
{
"files": {
".app.json": {
"sha256": "3bd8fab65d4855054391e5f42091c43f14eb0259f7f2a0ad86a0ac051dc0dd7e",
"size": 103
},
".codex-plugin/plugin.json": {
"sha256": "0052e36bdacfdaeb306a9f1ae6108a47eee24a4285cd698ae943ae18b64c0786",
"size": 7018
},
".mcp.json": {
"sha256": "fb8515d34e2a98b5f96c6fc2833ccfeaa94bf42b6a23b2c1f22c2b1c79f9a5ad",
"size": 133
},
"assets/icon.png": {
"sha256": "00dcd871473684d4e9449009ee90b1b1e0357b2ca6910314e18be09b8c6b6b7c",
"size": 22441
},
"assets/logo.png": {
"sha256": "00dcd871473684d4e9449009ee90b1b1e0357b2ca6910314e18be09b8c6b6b7c",
"size": 22441
},
"assets/screenshot-1-transcribe.png": {
"sha256": "2882cb107877183953e14fbd7d32b5f6e25e448fbff6d77df56d28aed2c81b82",
"size": 95871
},
"assets/screenshot-2-transcript.png": {
"sha256": "a557151d0ffa2270ca276de80b4b8e0d9930dc9daafbfafdd89d92b147609df3",
"size": 105478
},
"assets/screenshot-3-meetings.png": {
"sha256": "e7f66b64a9bade8795174164ba62aa82006c0f3d6d8be28f680422067db4c3d6",
"size": 107022
},
"assets/screenshot-4-quotes.png": {
"sha256": "d7548726886ec777c6e76f8428c730d9f2282ef498699733b1908eba4a457672",
"size": 102265
},
"assets/screenshot-5-edit.png": {
"sha256": "7a66baff1831350136e7eef6768f65f72271ffa9e02d60ee8f97e5e244acbef2",
"size": 90655
}
}
}SHA-256 of public snapshot: 4988bfb68bdbc412649c34bfb8fb086f19175eba82979f791551406bc12628f6