style-memory
Skill1 starsChanged 2 days agoMITView on GitHub
Brand style memory — tokens, component library, the video's manifest, golden references of successful videos, the memorable element, a switchable logo. Use when assembling any video and when the human says \"I like this one\".
In this repository23 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- ai-visualizationSkill
- approval-packSkill
- content-atomizationSkill
- dispatchSkill
- fact-checkSkill
- ffmpeg-skillSkill
- hook-storylineSkill
- idea-diceSkill
- jurySkill
- memory-compactionSkill
- perf-loopSkill
- publish-kitSkill
- seamless-insertSkill
- setupSkill
- shoot-briefSkill
- sound-designSkill
- source-skeletonSkill
- treg-gatewaySkill
- trend-scoutSkill
- utmSkill
- video-codeSkill
- weekly-cycleSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 33
When the human says "I like this one because…":
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Style memory
What is stored per client
clients/<c>/brand/
tokens.json # colors, fonts, radii, easing, durations, 9:16 safe zones
visual.md # style description + the MEMORABLE ELEMENT (the brand's recurring device)
templates/ # components: badge, callout, titles, transition, CTA, "AI visualization" badge, logo layer
references/ # external references the client likes (link + why)
golden/<name>/ # golden references: successful videos that later videos are built from
assets/ # logo SVG, fontsmanifest.json — for every video
Lives next to the final file. It lets you rebuild the video and repeat the style.
{"id":"W44-R03","flow":"f05-object-insert-inshot","duration":31.2,
"logo":{"enabled":true,"src":"brand/assets/logo.svg","position":"top-left"},
"tracks":[{"t0":0,"t1":2.1,"src":"a1b2c3d4e5f6@40.1","component":null},
{"t0":2.1,"t1":7.8,"src":"gen/flf_A_B.mp4","label_ai":true},
{"t0":0,"t1":31.2,"component":"subtitles/bold-yellow","data":"captions.json"}],
"generations":[{"step":"flf_video","model":"kling_o3_flf","prompt":"…","seed":123,"credits":40}],
"grade":{"lut":"brand/luts/warm.cube","strength":0.7,"grain":6},
"music":{"src":"…","license":"…"}, "stock":[]}Golden references
When the human says "I like this one because…":
- The feedback agent copies
manifest.json, prompts, models, seeds, presets and the preview intobrand/golden/<rubric>-<short-name>/. - It writes
why.mdin the human's own words: what exactly they like (pace, badge, transition, color). - In
visual.mdthe rubric links to the golden reference. The director builds the next video of this rubric from the golden reference: the same components, pace, grade.
Rules
- The logo is a separate layer, switched off by a flag without re-editing. For localizations — an export without text.
- AI B-roll is brought to the brand style: prompt prefix and references from
visual.md, a common LUT. - The memorable element is present in every video of the rubric. Without it, videos are not remembered.
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.

