The Agent Skills Directory

[Prompt Injection] (SAFE): The skill contains only instructional metadata and trigger descriptions. There are no attempts to bypass safety filters or override system instructions.
[Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file paths, or network exfiltration patterns were detected in the skill metadata.
[Obfuscation] (SAFE): No Base64, zero-width characters, or hidden unicode markers were found.
[Unverifiable Dependencies] (SAFE): No external package managers (pip, npm) or remote scripts are referenced.
[Indirect Prompt Injection] (SAFE): The skill defines a data ingestion surface (user requests for gitignores) and provides tool access (Bash, Write). However, without specific prompt templates provided in the file, there is no evidence of unsafe data interpolation or sanitization failure.
[No Code] (SAFE): This skill consists entirely of markdown metadata and documentation with no scripts or binary files attached.

gitignore-generator