mm-cli-skill

Indexes directories and extracts machine-readable metadata and content from multimodal assets via mm-ctx CLI commands.

16|Updated Jan 16, 2026
One-click install
npx skills add https://github.com/vlm-run/skills --skill mm-cli-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mm-cli-skill
Source: https://github.com/vlm-run/skills/tree/main/skills/mm-cli-skill
Command: npx skills add https://github.com/vlm-run/skills --skill mm-cli-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The mm CLI skill enables fast, deterministic exploration and extraction of content from multimodal directories, turning scattered files into searchable, structured insights.

Core Features & Use Cases

  • Directory indexing: quickly index and query images, videos, PDFs, code, and other files within a directory.
  • Content discovery: list by type, size, or path, and view directory trees for quick context.
  • LLM-assisted descriptions: generate summaries or metadata for complex documents and media (via mm's capabilities).

Quick Start

Start by inspecting a directory with the mm find command to reveal its structure and contents.

Frequently Asked Questions about mm-cli-skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I index and search for multimodal files in a directory?

To index and search multimodal directories, you use the mm CLI tool to scan images, videos, and PDFs, generating a structured, machine-readable index for rapid content discovery and metadata querying.

Can I extract text and metadata from images and PDFs using a CLI?

Yes, text extraction from images and PDFs is supported by the mm CLI, which applies deterministic directory indexing to parse file contents and expose structured metadata for downstream filtering.

Does the mm CLI require any external tools to analyze directory contents?

Yes, performing directory analysis with the mm CLI requires the mm-ctx tool alongside standard CLI commands to execute fast indexing, searching, and content extraction across multimodal assets.

What is the best way to generate LLM-assisted descriptions for scattered media files?

Generating LLM-assisted descriptions for media files is achieved by indexing the directory with the mm CLI, which leverages mm's capabilities to produce summaries and metadata for complex multimodal assets.

How do I filter directory trees by file type and size for multimodal assets?

Filtering directory trees by file type and size is handled by the mm find command, which lists multimodal assets based on specified attributes to provide immediate structural context for content discovery.

When should I use a deterministic CLI for directory analysis instead of manual file inspection?

A deterministic CLI for directory analysis is necessary when workflows demand rapid discovery and extraction across large volumes of images, videos, PDFs, and code, turning scattered files into searchable, structured insights.