100Critical
The skill allows arbitrary shell command execution and local file content
Behavior Analysis
Mismatch detected — The skill claims to 'generate infographics' but only prepares the content and creates a prompt for an infographic. The actual image generation is delegated to an external, unspecified image generation skill.
Claims to do
Infographic Generator: Two dimensions: **layout** (information structure) × **style** (visual aesthetics). Freely combine any layout with any style.
Actually does
This skill processes user-provided content (from a file or direct input) to generate a detailed prompt for an infographic. It reads local configuration from `EXTEND.md` and various internal reference files to analyze and structure the content. Finally, it calls an unspecified external image generation skill to produce an `infographic.png` file, performing local file operations for saving and backing up intermediate and final outputs.
Skill Info
Namejimliu/baoyu-skills/baoyu-infographic
Registrygithub
Versione13092ba2af1
PURLpkg:github/jimliu/baoyu-skills@e13092ba2af1?skill=baoyu-infographic
Stars14,596
Installs15,900
Source
SHA-256c81d8862997c...
SHA-1569e0550e8aa...
MD589cc10a2cdae...
TLSHbd4291b85c16...
ssdeep192:eZts5oXS...
Size12,909 bytes
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
Assessments (3)
AI Agent Traps ↗Local File Inclusion leading to Prompt Injectioncriticalprompt injection
The skill takes a file path as input (`path/to/content.md`) and uses user-controlled options (`--layout`, `--style`) to load content from `references/layouts/<layout>.md` and `references/styles/<style>.md`. This content, along with user-provided input, is combined to generate the final prompt for an image generation LLM, allowing arbitrary local file content to be injected into the LLM's prompt.
100% confidenceline 130
Combine: Layout definition from `references/layouts/<layout>.md` ... Structured content from Step 2
Configuration File Poisoning via EXTEND.mdhighsupply chain
The skill loads preferences from `EXTEND.md` files located in user-controlled directories (project, home). If an attacker can place a malicious `EXTEND.md` file, they could define custom styles or layouts containing prompt injection payloads or other malicious configurations that are then used in prompt generation.
90% confidenceline 153
test -f .baoyu-skills/baoyu-infographic/EXTEND.md && echo "project"
Command Execution Capabilityhighcommand execution
The skill explicitly uses shell commands (`test -f`, `echo`, `Test-Path`) to check for configuration files. This demonstrates the agent's ability to execute arbitrary shell commands, which could be exploited if user input is unsafely interpolated into commands.
100% confidenceline 153
test -f .baoyu-skills/baoyu-infographic/EXTEND.md && echo "project"
Unspecified Credential Stripping Mechanismlowdata exfiltration
The skill states it will 'strip any credentials, API keys, tokens, or secrets' from outputs. While this is a positive security intention, the mechanism is unspecified, and an ineffective implementation could still lead to sensitive data leakage.
80% confidenceline 139
strip any credentials, API keys, tokens, or secrets before including in outputs
Description Mismatchmediumdescription mismatch
The skill claims to 'generate infographics' but only prepares the content and creates a prompt for an infographic. The actual image generation is delegated to an external, unspecified image generation skill.
85% confidence
Step 6: Generate Image: '1. Select available image generation skill... 3. Call with prompt file and output path.'
Dependencies (1)
baoyu-skills.claude-plugin/marketplace.jsonnot scanned
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/github/jimliu/baoyu-skills/baoyu-infographic)HTML
<a href="https://mondoo.com/ai-agent-security/skills/github/jimliu/baoyu-skills/baoyu-infographic"><img src="https://mondoo.com/ai-agent-security/api/badge/github/jimliu/baoyu-skills/baoyu-infographic.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/github/jimliu/baoyu-skills/baoyu-infographic.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow