tl;dv automation · runs daily
Turn tl;dv key moments into clips and notes
After a discovery call, the moments that matter sit somewhere inside a 40-minute transcript, and nobody has time to cut them out.
Cotlet does it in your own tl;dv account with list_meetings, get_transcript, create_clip and add_note, called by your AI agent, an n8n workflow or a script.
Hosted · nothing to install or maintain
Can tl;dv do this on its own?
tl;dv's AI highlights moments by itself, and the API and the official MCP server read transcripts and notes only (checked 17 Oct 2026). The API highlights endpoint is deprecated in favor of notes, and the MCP docs say it cannot create or edit meetings. No official route creates a clip or adds a note by code.
How you do it in tl;dv today
- During a meeting, the pin icon flags a moment for the AI, and typing in the box before pinning adds a note; both are timestamped and stored with the recording.
- After the meeting, you select part of the transcript and click Add note, or click Create clip.
- From the recording, Create clip opens a pop-up with a slider for the segment and a field to rename the clip, and Save puts it in the Clips tab in the left sidebar.
- To share a clip, open it and press Copy link; anyone who gets the link can view it.
- Each clip is cut by hand, one at a time, so only the calls someone cared about get any, and the rest stay a long transcript.
How to turn tl;dv key moments into clips and notes
Read yesterday's Acme discovery call in tl;dv. Wherever the customer names a budget, a deadline or an objection, save that moment as a clip with a short title and add a timestamped note with the quote. - 01
Find the call
The agent picks the meeting named in your prompt from list_meetings, or yesterday's calls when you gave a rule, and waits for tl;dv to finish processing before reading anything.
- 02
Read the transcript for your triggers
get_transcript returns the call with timestamps, and the agent looks for the triggers you listed: a budget, a deadline, an objection. It keeps the exact time range around each one, from the sentence before to the sentence after.
- 03
Cut a clip per moment
create_clip saves each range to your Clips tab with a title built from the topic, for example "Budget confirmed for Q4". Moments closer than a minute are merged into one clip.
- 04
Pin the quote as a note
add_note attaches the speaker's own words at the same timestamp, so the note reads as evidence next to the AI notes. Nobody is alerted, because tl;dv sends Slack mentions only for notes added during a meeting.
- 05
Report links
The agent returns the clip titles with their times, so you decide which links to send. Creating a clip shares nothing.
What you get back
Acme discovery call, 15 Oct (42 min)
Clips saved: 3
- 00:12:05 to 00:12:53 "Budget confirmed for Q4"
- 00:27:40 to 00:28:10 "Needs SSO before launch"
- 00:35:02 to 00:35:30 "Worried about onboarding time"
Notes added: 3, each with the quote Other ways to ask
- After every sales call this week, clip each pricing question and add a note with who asked it.
- Find every moment in last week's customer calls where someone said "cancel" and clip it with the customer name in the title.
- Only add notes with the quotes, no clips, and send me the list.
- Clip the three decisions made in Monday's planning call and name each clip after its owner.
What tl;dv allows, and what to know first
- tl;dv's AI already groups a meeting into topics with a gist and timestamped notes, each linking to the point in the recording, and manual notes sit next to them.
- A note added after a meeting is edited or deleted from the three dots menu on the note, so a wrong quote is undone in the app.
- tl;dv's @mention alert in Slack works only for notes added during a meeting, so a note added afterwards sends no alert.
- Anyone who gets a clip link can view the clip, so copy a link only when the clip is safe for its reader.
- On the free plan, tl;dv limits AI highlighting and summarizing to the first 10 meetings.
create_clip and add_note change only your own tl;dv data and reach no one else, so they run without asking first.
Sources: tl;dv Help: Manually add timestamped notes to meetings · tl;dv Help: How to create, share, and download clips · tl;dv Help: AI Meeting Notes · tl;dv Public API docs (v1alpha1)
Call create_clip from a script
curl -X POST https://cotlet.dev/api/v1/tldv/meetings/mt_4f21a9/clips \
-H "Authorization: Bearer $COTLET_API_KEY" \
-H "Content-Type: application/json" \
-d '{"from_time":"00:12:05","to_time":"00:12:53","title":"Budget confirmed for Q4"}' {
"clip_id": "cl_71d0",
"meeting_id": "mt_4f21a9",
"title": "Budget confirmed for Q4",
"from_time": "00:12:05",
"to_time": "00:12:53",
"duration_seconds": 48,
"url": "https://tldv.io/app/clips/cl_71d0"
} MCP: add https://cotlet.dev/mcp/v1 with the same key; the call is tldv_write → create_clip. Example data.
Schedule it in n8n, Make, Zapier or crontl;dv tools this automation uses
| Tool | Parameters |
|---|---|
| list_meetings | query, from, to, folder_id, limit, cursor |
| get_transcript | meeting_id |
| create_clip | meeting_id, from_time, to_time, title |
| add_note | meeting_id, time, text |
Frequently asked questions
Is it safe to connect tl;dv to Cotlet?
Yes. You sign in to tl;dv yourself, in your own Cloud Browser in the EU, and we never save your password. Your agents only get tl;dv once you enable it, every background request is listed, and the browser profile is stored encrypted with AES-256-GCM, the cipher 1Password uses for its vaults. The AI model never sees a login, posts and messages wait for your OK, and Disable for agents cuts access in one click. Independent scanners grade cotlet.dev A+ on Qualys SSL Labs, securityheaders.com and Mozilla Observatory; live results are on https://cotlet.dev/security.
Give your agent access to tl;dv.
Free 7-day trial with 2,000 credits, no card needed. Cotlet is independent and not affiliated with tl;dv.