qwen-mm-plugins-example

Demonstrates MCP capability structure with demo tools for text, images, and API calls.

2.9k|185|Updated Jul 29, 2026
One-click install
npx skills add https://github.com/QwenLM/Qwen-MM-Plugins --skill qwen-mm-plugins-example-qwenlm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: qwen-mm-plugins-example
Source: https://github.com/QwenLM/Qwen-MM-Plugins/tree/main/src/capabilities/example/skill
Command: npx skills add https://github.com/QwenLM/Qwen-MM-Plugins --skill qwen-mm-plugins-example-qwenlm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Developers adding a new capability to Qwen-MM-Plugins need a working reference that shows how MCP tools, content-return shapes, API calls, and config resolution fit together, and this template provides that minimal working example. ## Core Features & Use Cases - Demo MCP tools: echo returns text, make_swatch generates a solid-color image, make_film_strip produces multi-frame image output, describe calls an OpenAI-compatible chat model, and config_probe reports resolved configuration. - Config resolution pattern: shows how shared.env.get_env resolves values with precedence of environment variable over ~/.qwen-mm-plugins/config over default. - Capability composition: demonstrates depending on qwen-mm-plugins-core tools like read_video to extract frames before doing capability-specific work. - Use Case: Copy the capabilities/example directory, adapt the demo tools into real ones, and follow the documented patterns to ship a new multimodal capability. ## Quick Start Ask the agent to run the config_probe tool from the example capability to see how environment and config values currently resolve.

Frequently Asked Questions about qwen-mm-plugins-example

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

FAQPage Schema
How do I add a new capability to Qwen-MM-Plugins?

Copy the capabilities/example directory and adapt it to your use case. The template includes demo MCP tools showing content-return shapes, API calls, and config access, and the repository's how-to-add-a-new-capability doc walks through the full process.

What demo tools does the example capability include?

It includes echo for text returns, make_swatch for single-image output, make_film_strip for multi-frame image output, describe for calling an OpenAI-compatible chat model, and config_probe for inspecting resolved configuration values.

How does config resolution work in Qwen-MM-Plugins?

Config resolves through shared.env.get_env with precedence of environment variable first, then the ~/.qwen-mm-plugins/config file, then the default value. Run the config_probe tool to see what currently resolves for your setup.

Can the describe tool run without an API key?

Yes, pass dry_run=true to the describe tool to see the request it would send without needing a key or network access. For real calls it uses DASHSCOPE_API_KEY resolved through the standard config chain.

How do capabilities handle video or document inputs?

The example capability has no media reader of its own, so it composes with qwen-mm-plugins-core. Call read_video from the core plugin to extract frames, then pass a frame path to tools like describe or make_swatch.