The Agent Skills Directory

[Remote Code Execution] (CRITICAL): The skill documentation explicitly instructs users to execute unverified remote code via curl -sSL https://canifi.com/skills/zendesk/install.sh | bash and curl -sSL https://canifi.com/install.sh | bash. This is a high-risk pattern that grants an untrusted domain full control over the local execution environment.
[External Downloads] (HIGH): The skill depends on installation and configuration scripts from canifi.com, which is not a recognized trusted repository or organization.
[Indirect Prompt Injection] (LOW): The skill acts on external data (support tickets and customer profiles) which can contain malicious instructions.
Ingestion points: Zendesk ticket content and customer profiles.
Boundary markers: No explicit markers or 'ignore embedded instructions' warnings are present in the documentation.
Capability inventory: The skill can create/modify tickets and search knowledge base articles.
Sanitization: No evidence of sanitization or validation of the external content before processing.

zendesk