The Agent Skills Directory

[SAFE]: The skill exhibits no malicious behavior. It functions as a specialized analysis engine that reads and reasons about architectural artifacts within a user-provided directory. While the skill ingests untrusted data during analysis, which is a surface for indirect prompt injection, this is a fundamental requirement of its primary purpose and is mitigated by the skill's focus on read-only analysis and reporting. The workflow does not include commands for unauthorized network access, data exfiltration, or privilege escalation. No hardcoded credentials, malicious persistence mechanisms, or obfuscated code were found.

explain-system-tradeoffs