The Agent Skills Directory

[Prompt Injection] (SAFE): The skill instructions in SKILL.md are purely descriptive and do not contain any attempt to bypass safety filters or override system instructions.
[Data Exposure & Exfiltration] (SAFE): No sensitive file access or network communication patterns were detected. The skill only processes provided parameters and returns structured data.
[Remote Code Execution] (SAFE): The implementation in skill.py does not use any dangerous functions like eval(), exec(), or subprocess. It relies on a local framework import which is standard for this type of skill.
[Indirect Prompt Injection] (SAFE):
Ingestion points: The skill ingests untrusted data through the source_content parameter.
Boundary markers: None present in the code; however, the skill has no dangerous capabilities to exploit.
Capability inventory: No file-write, network, or execution capabilities detected.
Sanitization: None present, but not required given the lack of exploitable capabilities.

learning-translation