Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/isaacsight/kernel/motion-prompt-languagenpx skills add isaacsight/kernel --skill motion-prompt-languagegit clone --depth 1 https://github.com/isaacsight/kernelWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/isaacsight/kernel/motion-prompt-language)<a href="https://agentmods.dev/skills/isaacsight/kernel/motion-prompt-language"><img src="https://agentmods.dev/badge/skills/isaacsight/kernel/motion-prompt-language.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00090 | $0.01126 |
| Opus 5 | $0.00045 | $0.00563 |
| Sonnet 5 | $0.00018 | $0.00225 |
| Haiku 4.5 | $0.00009 | $0.00113 |
Grade A, and why
motion-prompt-language scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Motion Prompt Language
Write generation prompts like a creative director, not a text-to-image prompt. Think in layers, in this order:
Concept → Visual language → Camera → Animation → Typography → Lighting → Texture → Timing
A prompt that specifies every layer produces motion closer to a design-studio reel than a typical AI animation. A prompt that skips layers gets generic defaults for the missing ones.
The layer stack
| Layer | What to specify | Example fragment |
|---|---|---|
| Concept | One-sentence premise | "A futuristic literary magazine comes to life." |
| Visual language | Style anchors + studio references | "Minimal Swiss design, Buck Design, DIA Studio" |
| Camera | Movement verbs, restraint | "Slow dolly, gentle zooms, no aggressive movement" |
| Animation | How elements move | "Elegant easing, kinetic typography, subtle parallax" |
| Typography | Role of type in the frame | "Large editorial typography; letters build from fragments" |
| Lighting | Source + mood | "Soft natural daylight, realistic shadows" |
| Texture | Material world | "Matte paper, visible fibers, ink bleed, Super 16mm grain" |
| Timing | Duration + pacing | "8 seconds. Everything moves deliberately." |
Quality vocabulary
Descriptors that consistently raise output quality:
- Studios (style anchors): Buck Design, Territory Studio, DIA Studio, ManvsMachine, Ordinary Folk, Instrument, Elastic; Refik Anadol for data-driven work; Saul Bass for kinetic type; Kurzgesagt for explainers
- Motion: editorial motion design, kinetic typography, procedural animation, elegant easing, subtle parallax, physics-based motion, cinematic camera movement
- Composition: layered compositions, negative space, modular grid system, Swiss typography, restrained color palette
- Surface: matte materials, paper texture, film grain, Super 16 aesthetic, analog imperfections, volumetric lighting
Useful genre templates
Kinetic typography — large editorial type fills the screen; letters break apart; sentences build from fragments; strong rhythm; black and cream; Saul Bass / Buck / DIA Studio.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 115 lines · 90 tokens per session scan A 106d3db4f1f6
motion-prompt-language is a skill published in the GitHub repository isaacsight/kernel (16 stars, last pushed 6d ago), licensed MIT. It adds 90 tokens to every session and 1,126 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
animejs-scrollcraft
Build animejs.com-grade scroll-driven kinetic pages with the vendored anime.js v4 bundle (website/vendor/anime.esm.min.js). Use when working on the prime-silo marketing site (website/) or any scrollytelling/animation surface. Contains the VERIFIED v4 API, the design language to emulate, the Prime-Silo set-piece specs…
delivery-board
Operate the Prime-Silo delivery board (delivery/) autonomously — select, claim, execute, verify, and log work contracts without frontier supervision. Use whenever picking up a task from delivery/board/BOARD.md, authoring a gate script, verifying another agent's task, or deciding which model tier should take which…
devops-pipeline
Build, test-package, and release the Prime-Silo desktop app (Space Agent). Use when asked to build the app, make a local/test build, package for Windows/macOS/Linux, cut a release, bump the version, or debug the release CI. Covers the Open-Studio companion services (opencode + open-notebook) bundling.
house-trainer
Operate the EP-T house-method QLoRA trainer end-to-end on the T480 eGPU — dataset builds (T2), training/eval/GGUF (T3+), verification, and the data-quality levers. Use when training, evaluating, rebuilding the dataset, verifying a T-task, or extending Workstream T.
Benny pilot
Be the operator's grounded co-pilot inside the Bridge cockpit (#/primesilo/bridge). Load this when the user is on the Bridge and asks "what am I looking at", "explain this graph", "what did I work on", "ingest these docs", "re-run the last manifest", or any question about the mesh on screen — memory, documents, code…
prime-silo-experience
The Prime-Silo product experience contract — identity (Prime-Silo + Agent Benny, never Space Agent), earth-tone design tokens, progressive-discovery doctrine, and the "motion is meaning" choreography rules that bring animejs.com-grade kinetics INTO the app without violating the C0 calm contract. Use for any…