The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data from software repositories to detect piracy and malware.
Ingestion points: The analyze_suspicious_repo and is_piracy_repository functions in SKILL.md take external content such as readme_content and repository topics as input for evaluation.
Boundary markers: There are no explicit boundary markers or "ignore instructions" directives used when interpolating or processing these external data strings.
Capability inventory: The skill uses regular expression matching and logical scoring to evaluate input. It does not invoke high-risk tools or system-level commands based on the data analysis.
Sanitization: The provided logic does not include sanitization, filtering, or escaping of the external strings before they are processed by the analysis functions.

marvelous-designer-unlock-analysis