The Agent Skills Directory

PROMPT_INJECTION (LOW): The skill exhibits susceptibility to indirect prompt injection via external data ingestion. 1. Ingestion points: The scripts brand_voice_analyzer.py and seo_optimizer.py process external files such as existing_content.txt and blog_post.md. 2. Boundary markers: Absent; there are no specified delimiters or 'ignore' instructions to prevent the agent from following commands embedded within the analyzed content. 3. Capability inventory: The skill uses subprocess calls to execute local Python scripts and shell utilities like grep. 4. Sanitization: Absent; no sanitization or escaping of the input file content is described before it is passed to the scripts.
COMMAND_EXECUTION (SAFE): The usage of local Python scripts and basic shell commands (grep, cp) is restricted to the skill's local directory and intended marketing workflow, posing no unusual risk.

content-creator