The Agent Skills Directory

[General Security] (SAFE): The skill is purely instructional and contains no evidence of malicious behavior, obfuscation, or unauthorized access to system resources.- [Indirect Prompt Injection] (LOW): The code_context parameter allows for the ingestion of external data. This is a standard feature for coding assistant skills. Since the skill lacks the capability to execute commands, modify the file system, or exfiltrate data, the risk is limited to influencing the agent's internal reasoning or output quality. No specific boundary markers or sanitization logic are defined in the provided file, but the lack of dangerous tools mitigates the threat.

fundamentals