The Agent Skills Directory

[SAFE]: The skill provides legitimate security audit functionality through scripts like vulnerability_scanner.py and dependency_auditor.py, which are designed to detect security weaknesses in target applications.
[SAFE]: The attack_patterns.md file contains a library of security test payloads (e.g., for XSS, SQL injection, and SSRF) which are appropriately documented as examples for use during authorized security assessments.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data such as source code files and package manifests. While this presents an ingestion surface for untrusted data, the skill's capabilities are restricted to scanning and reporting within the local environment, and it does not perform high-risk operations like network exfiltration or shell command execution on the processed data.
[SAFE]: The skill follows industry best practices by including a responsible disclosure guide (responsible_disclosure.md) and emphasizing the necessity of written authorization before performing any security testing.

security-pen-testing