The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill is explicitly designed to handle untrusted external data. While it uses robust schema validation (Zod), it processes free-text fields like titles using regex. If the output of this skill (e.g., quality warnings or the data itself) is fed into a downstream LLM without sanitization, it could serve as a vector for indirect injection. However, the skill's purpose is inherently defensive.
[External Downloads] (LOW): The implementation depends on the zod package. This is a highly trusted, standard validation library used widely in the TypeScript ecosystem.
[No Malicious Patterns] (SAFE): No evidence of prompt injection, data exfiltration, obfuscation, or unauthorized command execution was found in the code or metadata.

validation-quarantine