Use Claude Code with Tareno to turn the release notes of a user-facing feature into a 45-second Shorts script and title and a draft for your approval in Tareno.
Can Claude Code prepare a YouTube Short for a new feature from my repository?
Partly. Claude Code reads the release notes and the changed interface code, then writes a 45-second script and a title of up to 100 characters. It saves a draft through Tareno once your screen recording is in the media library. It cannot record or cut the video, and scheduling stays pending until you approve it in Tareno.
A web product's best release announcement is often the feature itself, shown for forty seconds with a calm voice. Product teams know this and still ship without it, because the script, the recording and the upload add an afternoon to a release day that is already full. Claude Code can take the writing off that afternoon: it reads the release notes and the interface code, knows what changed on screen and in what order, and turns that into a spoken script. This page follows a team that ships a dashboard filter, with the one rule that matters most for UI recordings: customer data must never be visible in the frame.
Example: a reporting dashboard ships saved filters
Imagine a team whose reporting dashboard now lets users save filter combinations. The product engineer asks Claude Code to read the release notes and the pull request for the filter panel, and to name the three steps a user takes: set a filter, name it, reuse it. She records the flow in a demo workspace with fake data, exports it vertically at 1080 by 1920 and adds it to the Tareno library. Claude Code writes a 45-second script whose sentences match the three steps and proposes three titles under 100 characters. It finds the file with list_media, saves the draft through create_drafts and returns the draft ID. After watching the clip next to the script in Tareno she requests Wednesday 11:00 in Europe/London, which waits as a pending action.
Step by step
01
Ask for the user's steps, not the commits
Have Claude Code read the release notes and the interface code, then list what a user does with the feature in order. A Short follows a user's steps; it does not follow the order of the pull requests.
02
Record in a demo workspace
Use fake data in a workspace made for demos, so no customer name, email or revenue figure can appear. Export a vertical clip of up to 60 seconds and add it to the Tareno library.
03
Write the script to the footage
Ask for a spoken script of about 45 seconds whose sentences match the on-screen steps, and three titles of at most 100 characters. Tell Claude Code the length of your recording so the pace fits.
04
Confirm the feature is live
Ask Claude Code to check in the code that the feature is enabled for all users, not behind a flag or a plan. A Short about a feature viewers cannot find creates support requests.
05
Save the draft
Claude Code finds the recording with list_media and saves title and file through create_drafts for your channel. Watch the clip next to the script in Tareno before you ask for a time.
06
Request the slot and approve it
Give a date, a time and an IANA time zone. The request becomes a pending action; approve it in the dashboard, and ask Claude Code to check it with get_action_status afterwards.
Prompts to copy
List the user's steps
Read the release notes for [saved filters] and the code of the filter panel. List in order the steps a user takes to use the feature, in plain words, and tell me which parts of the interface are visible at each step. Do not write the script yet.
Script and titles
I recorded [saved-filters-demo.mp4], a 40-second clip of those steps with demo data. Write a spoken script of about 45 seconds whose sentences match the three steps in order, and give me three titles of at most 100 characters with the character count.
Check that it is live
Check in the code whether the saved-filters feature is enabled for every plan and every user, or whether a flag or a plan limit hides it. Tell me exactly what restricts it, so I can say so in the Short or wait with the release. Do not call Tareno until I have answered.
Save and request
Use Tareno to find the file [saved-filters-demo.mp4] and save it as a YouTube draft for [channel name] with title option [2]. After I confirm, ask Tareno to schedule it for [Wednesday 11:00] in the [Europe/London] time zone and return the pending action ID.
YouTube limits and tools in Tareno
Text
up to 100 characters; media is required, text alone is rejected; plain text, no rich formatting
Images
jpg, png, up to 2 MB each, up to 1 per post, recommended ratios 9:16
Video
mp4, mov, 1 to 60 s, up to 50 MB, one video per post, ratio 9:16
Links
links in the text are not supported
What Tareno can read
posts published outside Tareno: yes; comments: yes; transcripts: yes; video download: yes
As configured in Tareno on 2026-10-24. The live rules for your own account come from the get_platform_schema tool.
What YouTube itself documents
YouTube allows Shorts of up to three minutes in vertical format. Tareno's configuration for YouTube currently stops at 60 seconds of vertical video, so a longer clip has to be shortened first.
Source: YouTube Help: YouTube Shorts, read on 2026-10-24. Platform rules change; check the page before you rely on a number.
YouTube in Tareno
A feature Short carries one idea, which suits the single vertical video per post that YouTube takes in Tareno; a feature with three parts becomes three Shorts, not one long one.
A UI recording shows more than the feature: tabs, notifications and names appear at the edges, so a demo workspace with fake data is safer than blurring afterwards.
Tareno's YouTube text holds a title and little else, so what the feature costs or which plan includes it belongs in the script or a pinned comment, not in the saved text.
Comments on published Shorts can be read through Tareno, so Claude Code can summarize what users asked about the feature and feed the next release note.
What does not work
Claude Code cannot record or edit the interface demo, so the video must exist in your Tareno media library before a draft can be saved.
It cannot judge whether a visible name or number in a recording is private, so a person checks every frame before upload.
It cannot publish the Short by itself, since scheduling and publishing requests wait for your approval in the Tareno dashboard.
Mistakes to avoid
Recording with real data. Even a blurred customer name can be recovered or guessed, so use a demo workspace.
Announcing a feature that is plan-limited without saying so. State who gets it in the script, or the comments will do it for you.
Reading the pull request title as the script. Titles describe code changes; the script should describe what the user can now do.
Skipping the live check. A flag that is still off turns a release Short into a promise you cannot keep.
Open the published Tareno listing, choose install or connect, and complete the Tareno login. Enable Tareno in a new conversation. For Claude, search for Tareno in the supported plugin directory.
Choose OAuth, connect and sign in with your own Tareno account. Check the requested permissions. Refresh the client's tool list after authorization.
Run list_accounts first. Confirm the returned workspace and account IDs before creating a draft. If no accounts appear, connect a social account in Tareno first.
https://tareno.co/api/mcp
Use the same Tareno account across supported clients. Open the relevant setup guide for authentication, first use, reconnect, and revocation.
Can Claude Code write a script for a feature video?
Yes. It reads the release notes and the interface code, lists the steps a user takes and writes a spoken script that follows them. You record the clip, and Claude Code pairs it with the title in your Tareno media library.
Why record in a demo workspace?
A UI recording shows names, numbers and notifications at the edges, and blurring afterwards is unreliable. A workspace with fake data removes the risk before the clip exists, and Claude Code only ever sees the file you upload.
How can Claude Code know the feature is live?
It can check the code for flags and plan limits that restrict the feature and tell you what it found. Treat that as a strong hint and confirm in production, because configuration outside the repository can differ.
How long should a feature Short be?
Between 30 and 45 seconds works for one idea. Tareno's configuration lists 1 to 60 seconds for YouTube, so a script of 45 seconds leaves room for a calm pace and a closing line.
Where does the feature link go?
Into a pinned comment or the description in YouTube Studio. The text Tareno saves for YouTube is limited to 100 characters and does not support links, so it holds the title and nothing else.
Will the Short publish without my approval?
No. A scheduling request becomes a pending action in Tareno. YouTube receives the video only after you approve it in the dashboard, and `get_action_status` shows what happened afterwards.