- Home
- Skills
- Image & Video
- Video Fetcher: Any Video, Transcript or Archive Note
More screenshots
Works with the AI Agents you already use
Video Fetcher: Any Video, Transcript or Archive Note
Turn any video into a transcript, a summary, or an archived note — captions first when they exist, local transcription when they don't, frames read alongside the words.
$5
Video Fetcher: Any Video, Transcript or Archive Note
Example session with this skill installed
A friend sent this 6-minute TikTok on sourdough starters — what does it say, and save it so I can find it later.
- Read your context and instructions
- Compiled the video fetcher
Path picked first: no captions exist on this short video, so the captions path was skipped openly and the local transcription path was used instead — the switch was stated, not silent. Downloaded the clip and transcribed it locally with Whisper, with the domain vocabulary supplied as hints (starter, discard, hydration, bulk ferment) so the tool names and baking terms came through clean. The transcript was read with a skeptical eye: one passage over background music was flagged as suspect rather than quoted, and a misheard tool name was corrected only because the frames confirmed it. Frames were read alongside the speech — a contact sheet of evenly spaced frames, each labeled with its timestamp, showed the feeding schedule written on screen, which the spoken words never state outright. The archive note carries the title, source, creator, upload date, and duration; a 4-sentence summary in plain words; the full transcript in its own section; timestamped key moments, including the on-screen feeding schedule; the contact sheet inline; and provenance — where it came from, when it was fetched, and which path was used. The saved note was reported with its path, and the one-line answer — a 1:1:1 feeding ratio on a 12-hour schedule — came with the timestamp of the moment it is stated.
Connects securely to your tools. The creator never sees your data.
About this skill
Someone sends you a video link and asks what it says, or save this for later. This skill is the decision tree and the tooling for both, across platforms — video services and local files alike. It picks the fast captions path when captions exist and local transcription when they do not, matches the output to the request instead of doing all of them, reads the frames as well as the speech, and archives the result as a Markdown note with provenance. Use it whenever a video arrives as a link and the answer has to be text you can search, quote, and find again.
It reads public videos and files you supply; login-walled sources may refuse anonymous downloads, and the skill says so instead of fighting the wall.
Machine captions and Whisper output both mishear names and jargon — corrections are made only when the frames or the user confirm them.
Music sections produce hallucinated lyrics; any transcript over background music is treated as suspect.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 days ago
Permissions
No special permissions declared or detected