The Agent Skills Directory

[SAFE]: The skill provides a comprehensive library of patterns for structural code analysis. All identified patterns are used for detecting issues in external codebases and do not contain malicious instructions for the agent itself.
[EXTERNAL_DOWNLOADS]: The skill references documentation and resources from well-known and trusted sources, including ast-grep.github.io, owasp.org, and clean-code-developer.com.
[COMMAND_EXECUTION]: Uses standard utilities like ast-grep (sg) and ripgrep (rg) for searching and analyzing code. The commands are appropriately scoped to the skill's primary purpose of code auditing.
[DATA_EXFILTRATION]: No network exfiltration or unauthorized file access patterns were found. Patterns targeting credentials (e.g., apiKey, password) are strictly for detection within the user's codebase and use safe placeholders.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external code for analysis, which represents an inherent surface for indirect prompt injection. However, the risk is minimal as the skill's functionality is limited to structural matching and reporting findings without executing the analyzed code.

code-antipatterns-analysis