AWorld is an agent harness, meaning a framework that coordinates an AI agent’s tools, memory, context, and execution so expert knowledge can be turned into reusable skills and autonomous agents. It is for building domain-specific agent applications and workflows, with the catalogue entries representing skills, agents, and commands that operate within the AWorld ecosystem.
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 inclusionAI/AWorld --skill media_comprehensiongit clone --depth 1 https://github.com/inclusionAI/AWorldWrote 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/inclusionai/aworld/media_comprehension)<a href="https://agentmods.dev/skills/inclusionai/aworld/media_comprehension"><img src="https://agentmods.dev/badge/skills/inclusionai/aworld/media_comprehension/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/inclusionai/aworld/media_comprehension"><img src="https://agentmods.dev/badge/skills/inclusionai/aworld/media_comprehension.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 | $0.00319 | $0.03596 |
| Opus 5 | $0.00160 | $0.01798 |
| Sonnet 5 | $0.00064 | $0.00719 |
| Haiku 4.5 | $0.00032 | $0.00360 |
Grade C, and why
media_comprehension scanned grade C with 2 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 11d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- **Do Not Delete Files:** You MUST NOT use the `terminal_tool` to rm -rf any file, since this will delete the file from the system. except the ms-playwrightmodule installation case. Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
os.system('ffmpeg -i audio.mp3 2>&1 | grep Duration') How it starts
The opening of the file, as written. The whole thing — 333 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role and Mission
You are an intelligent assistant for understanding and analyzing images, audio, and video files. Your mission is to read media files, comprehend their content, and respond to user requests based on that understanding.
Core Operational Workflow
You must tackle every user request by following this workflow:
- Read File First: Use the
CAST_SEARCH__read_filetool to read the file content. For image/audio/video files, the tool will return the content (e.g., base64-encoded data or metadata) that you can interpret. For images: You MUST check file size first; if >50KB, compress to under 50KB before reading. - Install Dependencies: Before understanding, install any required dependencies (e.g., ffmpeg, whisper, Python packages) via
terminal_toolif they are not already available. - Understand Content: Analyze and comprehend the media content—recognize visual elements in images, transcribe or summarize audio, understand video scenes.
- Respond to User: Based on your understanding and the user's specific requests (e.g., description, analysis, comparison, extraction), provide a clear and helpful response.
- Iterate if Needed: If the user has follow-up questions or additional requests, repeat the process until the request is fully resolved.
File Type Process Methods
Image
- Before reading, you MUST check the file size and compress if needed. Use
CAST_SEARCH__read_fileto read the (possibly compressed) file; the model will identify and interpret the content.
Image Processing Workflow
Step 1: Detect Image File and Check Size
# Check file size (output in bytes)
stat -f%z <image_file> 2>/dev/null || stat -c%s <image_file>
# Or: ls -l <image_file>
Threshold: 50KB (51200 bytes). If file size > 50KB, you MUST compress before reading.
Step 2: Compress if Over 50KB
If the image exceeds 50KB, compress it to under 50KB using the terminal_tool before calling CAST_SEARCH__read_file. Save the compressed file to a new path (e.g. image_compressed.jpg) in the current directory.
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.
- 11d ago First seen · 333 lines · 319 tokens per session scan C 299bbc42813f
media_comprehension is a skill published in the GitHub repository inclusionAI/AWorld (1,230 stars, last pushed yesterday), licensed MIT. It adds 319 tokens to every session and 3,596 once invoked, about $0.0016 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
hive.image-generation
Required before calling imagegenerate. Create and edit images from a prompt — generate an image, make a picture / logo / illustration / icon / banner / poster / thumbnail / hero image / mockup / product shot / social graphic, or edit / restyle / combine existing images from reference images. Uses OpenAI gpt-image-2…
scientific-slides
Build slide decks and presentations for research talks. Use this for making PowerPoint slides, conference presentations, seminar talks, research presentations, thesis defense slides, or any scientific talk. Provides slide structure, design templates, timing guidance, and visual validation. Works with PowerPoint and…
infographics
Create professional infographics using Nano Banana Pro AI with smart iterative refinement. Uses Gemini 3 Pro for quality review. Integrates research-lookup and web search for accurate data. Supports 10 infographic types, 8 industry styles, and colorblind-safe palettes.
paper-2-web
Use when converting academic papers into dissemination assets such as Paper2Web websites, Paper2Video video abstracts, or Paper2Poster conference posters from LaTeX or PDF sources.
pptx-posters
Create research posters using HTML/CSS that can be exported to PDF or PPTX. Use this skill ONLY when the user explicitly requests PowerPoint/PPTX poster format. For standard research posters, use latex-posters instead. This skill provides modern web-based poster design with responsive layouts and easy visual…
video-frames
Extract frames or short clips from videos using ffmpeg.