The Agent Skills Directory

[Metadata Poisoning] (SAFE): The skill metadata (name, description) is accurate and does not contain malicious instructions or deceptive claims.
[Prompt Injection] (SAFE): No malicious instructions, bypass markers, or jailbreak attempts were found in the skill content.
[Data Exposure & Exfiltration] (SAFE): No sensitive file paths, hardcoded credentials, or network-capable tools are utilized.
[Indirect Prompt Injection] (SAFE): The skill has a surface for indirect prompt injection via the processing of local source files.
Ingestion points: Read, Glob, and Grep tools applied to project files.
Boundary markers: None present in the instructions.
Capability inventory: Write and Edit file system tools.
Sanitization: No input validation or sanitization is defined.
Context: The risk is evaluated as SAFE because the toolset lacks command execution or network access.

dojo-test