The Agent Skills Directory

[SAFE]: No malicious patterns or vulnerabilities were detected across the analyzed files. The skill content consists of pedagogical security materials and standard engineering workflows.
[PROMPT_INJECTION]: The skill utilizes role-play instructions to adopt a "Senior Security Engineer" persona, but it does not contain attempts to override core safety guidelines, bypass restrictions, or extract system prompts. It explicitly prohibits the generation of functional exploit code without authorization.
[DATA_EXFILTRATION]: The skill includes regular expressions and logic for detecting secrets (e.g., AWS keys, GitHub tokens) as part of a security auditing toolkit. No hardcoded credentials or logic for exfiltrating discovered data to external domains were identified.
[COMMAND_EXECUTION]: While the skill provides Python code snippets for security scanning and cryptographic operations, these are provided as reference implementation patterns. There are no instructions for the agent to execute arbitrary shell commands or perform unauthorized system modifications.
[EXTERNAL_DOWNLOADS]: The skill references established industry resources such as the OWASP Top 10 project and mentions standard security testing tools. No suspicious remote code execution patterns (e.g., curl | bash) or downloads from untrusted sources are present.

senior-security