Oathe Security Badge

Is baagad-ai/content-wand safe?

https://github.com/baagad-ai/content-wand

90
SAFE

This is a well-designed content transformation skill with comprehensive security considerations including injection detection, trust boundaries, and URL validation. While external content processing presents some risk, the defensive measures are thorough and the skill operates purely through prompts without executable code.

Category Scores

Prompt Injection 85/100 · 30%
Data Exfiltration 90/100 · 25%
Code Execution 95/100 · 20%
Clone Behavior 95/100 · 10%
Canary Integrity 100/100 · 10%
Behavioral Reasoning 80/100 · 5%

Findings (5)

MEDIUM Complex External Content Processing -15 ▶

The skill processes external content from URLs and web searches, creating potential injection vectors despite comprehensive detection mechanisms

LOW External Network Access -10 ▶

Skill can fetch arbitrary URLs (with validation) which could theoretically be used for data exfiltration, though proper safeguards are in place

LOW Shell Script Present -5 ▶

Repository contains one executable shell script for health checking

LOW Standard Git Clone Operation -5 ▶

Normal git clone behavior observed with no anomalies

LOW Multi-Step Pipeline Complexity -20 ▶

Complex hub-spoke architecture with multiple sub-skills increases potential attack surface