sports-highlights
Skill1 starsChanged 9 days agoMITView on GitHub
Make a sports or fitness video with Vidova, the AI video editor. Use when someone asks for match highlights, a player edit, a game day video, a training or gym reel, a race or a season recap. Vidova cuts, captions, animates, adds music and renders the video in the cloud.
In this repository11 other files it gives its agents.
- beat-sync-editSkill
- birthday-videoSkill
- event-recap-videoSkill
- explainer-videoSkill
- family-videoSkill
- motion-graphicsSkill
- music-videoSkill
- podcast-clipsSkill
- product-ad-videoSkill
- travel-recap-videoSkill
- vidovaSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 4 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
Check that the Vidova tools are available, for example with get_active_project.
- Very specific: line 16
If the server offers a routine for a new edit from uploads, start it and follow its handoff.
- Very specific: line 19
Call render_video, wait for it with wait_for_jobs, then give the person the project url and the link to the finished video.
- Distinctive: line 25
Keep the real sound of the strike or the net on the moment.
- Distinctive: line 27
Commit to the slow part. A timid slow-down reads as a stutter.
- Distinctive: line 28
Never crop a head or the ball out of the frame, and never warp a body.
- Distinctive: line 29
Use only their names, numbers, scores, teams and times.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Sports and fitness
Sport is moments: the release, the strike, the finish, the last rep. A good edit finds the exact frame of each one and lands it on the song's hit.
Steps
- Check that the Vidova tools are available, for example with
get_active_project. If they are not, follow thevidovaskill to connect. - Make a project with
create_project. Bring the files in withimport_asset_urlfor a link,begin_asset_uploadfor a local file when you can run shell commands, orupload_filesto give the person the link where they add files. Callwait_for_assetuntil the files are ready. - Find each clip's peak to a tenth of a second with
read_asset,read_analysis_rangeandpreview_frame. Usepin_momentto keep it. Read the song withdetect_beats. - If the server offers a routine for a new edit from uploads, start it and follow its handoff. Otherwise call
search_templates. When a template fits, useread_templateandapply_template, then adjust it withpatch_film. When none fits, build the film withpatch_film. - Build each moment in three shots: the approach rushing in, the peak on a downbeat easing to about quarter speed, then the replay from another angle.
- Watch the result with
preview_timeline, and withpreview_rangeacross the moments that move. Callcheck_edit, thenwait_for_jobswith its Job ID, and fix what it names. - Call
render_video, wait for it withwait_for_jobs, then give the person the projecturland the link to the finished video.
What makes it work
- The person's brief comes first: their order, song, length, format and rules. This skill fills in the rest.
- Open as the brief says. When it says nothing, open on the best moment or on the athlete's face.
- Keep the real sound of the strike or the net on the moment. Let the music dip for half a second before it, then come back on the next downbeat.
- End on the finish: the celebration, the touch at the wall, the last rep or a face, slowed so it still moves.
- Commit to the slow part. A timid slow-down reads as a stutter. Keep the action moving and never freeze it.
- Never crop a head or the ball out of the frame, and never warp a body.
- Use only their names, numbers, scores, teams and times. Never invent a score or a statistic.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

