The Agent Skills Directory

[SAFE]: The skill instructions provide a comprehensive methodology for security assessments with a primary focus on authorization, rules of engagement, and safe testing practices. It explicitly forbids testing production systems, using real secrets, or operating without verified permission.
[SAFE]: The included Python script score_campaign.py is a data validation and reporting utility. It processes local JSON files to summarize campaign results and calculate risk metrics. It does not perform network operations, execute arbitrary code, or access sensitive system paths.
[SAFE]: External links in the taxonomy document reference established security and standards organizations such as OWASP, NIST, and MITRE. These are well-known and reputable sources for security research and best practices.
[SAFE]: The skill workflow incorporates benign canary values and mock oracles to ensure that security testing does not lead to unintended external effects or data exposure.

agent-red-teaming