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 skills add madebyaris/advance-minimax-m3-cursor-rules --skill minimax-m3-multimodal-inputgit clone --depth 1 https://github.com/madebyaris/advance-minimax-m3-cursor-rulesWrote 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/madebyaris/advance-minimax-m3-cursor-rules/minimax-m3-multimodal-input)<a href="https://agentmods.dev/skills/madebyaris/advance-minimax-m3-cursor-rules/minimax-m3-multimodal-input"><img src="https://agentmods.dev/badge/skills/madebyaris/advance-minimax-m3-cursor-rules/minimax-m3-multimodal-input/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/madebyaris/advance-minimax-m3-cursor-rules/minimax-m3-multimodal-input"><img src="https://agentmods.dev/badge/skills/madebyaris/advance-minimax-m3-cursor-rules/minimax-m3-multimodal-input.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 5.1 | $0.00121 | $0.01499 |
| Opus 5 | $0.00060 | $0.00749 |
| Sonnet 5 | $0.00024 | $0.00300 |
| Haiku 4.5 | $0.00012 | $0.00150 |
Grade A, and why
minimax-m3-multimodal-input 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 13d ago.
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 — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
M3 Multimodal Input
M3 accepts text + image + video as native input. The point of this skill is to use that capability honestly: ground every visual claim in the actual file the model can read, and re-read the post-change state before declaring a visual fix done.
When to Use
- The user attaches an image, screenshot, mockup, frame, or short clip.
- The task involves "make it look like this", "match this design", "why does this UI look wrong", "read this error screenshot", or "match this reference video".
- You are about to claim a visual or styling result. Any visual claim without a
multimodal-groundedread is a guess. - A bug report or feature request references a UI state the user can show you (rather than describe in words).
If the user is only talking about generating media (creating new images, video, TTS, music), the minimax-multimodal-toolkit skill is the right one — that skill is for output; this one is for input.
Step 0: Inventory The Input
Before reasoning, identify what the user attached and what you can actually read:
- Static image (PNG / JPG / WebP / SVG) —
Readthe file path the user provided or that the runtime surfaces. - Multi-frame video or screen recording — pick a small number of representative frames (start, mid, end) and reason about the in-between motion explicitly.
- Inline attachment that the runtime renders into the chat — the model sees it directly; cite it as "the attached image" with the visible region.
- Multiple images in a set (desktop + tablet + mobile mockups, before + after screenshots) — re-read each before claiming a responsive match or a fix.
If you cannot actually see the file, say so and ask. Do not invent the contents of an image you did not open.
Step 1: Ground In The File, Not The Prose
Always Read the file/frame and reference exact paths. Do not paraphrase a guessed description.
- Cite the file path (
/path/to/screenshot.png) and, when relevant, the region (top-right nav,hero block,error toast). - Quote visible text directly — error messages, button labels, empty-state copy. Do not paraphrase.
- If the user described the image in prose, treat their prose as a hint, not as evidence. The image is the evidence.
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.
- 13d ago First seen · 114 lines · 121 tokens per session scan A e9cbf96e3745
minimax-m3-multimodal-input is a skill published in the GitHub repository madebyaris/advance-minimax-m3-cursor-rules (125 stars, last pushed 2mo ago), licensed MIT. It adds 121 tokens to every session and 1,499 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
design-review
Stress-test a plan or design against domain language, docs, and code. Ask hard questions one at a time. Update CONTEXT.md and ADRs as decisions crystallize. Use when you already have a draft plan or spec and want to find gaps before implementation.
spark-environment-setup
Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.
spark-training-gotchas
Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.
motion-graphics
Use when the user wants a short, design-led motion graphic where motion is the message: kinetic typography, stat or number count-up, chart/data-viz hit, logo sting, brand lockup, lower-third, callout, social overlay, animated headline/tweet/news item, motion poster, or quick captured-page highlight. Usually under 10s…
dashscope
DashScope (Alibaba Cloud Bailian / 阿里云百炼) integration — image generation (qwen-image-2.0-pro), text-to-speech (qwen3-tts-flash), and ASR with word-level timestamps (qwen3-asr-flash-filetrans). Use when generating images via Qwen-Image, narrating via Qwen-TTS, or transcribing with word-level timestamps via Qwen-ASR.
vercel-composition-patterns
React composition patterns that scale. Use when refactoring components with boolean prop proliferation, building flexible component libraries, or designing reusable APIs. Triggers on tasks involving compound components, render props, context providers, or component architecture. Includes React 19 API changes.