The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/security_suite.py uses the subprocess module to execute system utilities and user-provided binaries.
Evidence: Functions _run and shutil_which invoke system tools such as file, otool, ldd, strings, ps, and lsof to collect metadata and runtime traces.
Context: This execution is central to the skill's primary purpose as a security analysis tool and is performed within a designated sandbox directory structure.
[PROMPT_INJECTION]: The skill includes a red-teaming component designed to detect prompt injection vulnerabilities in other repository files.
Evidence: references/agentops-redteam-pack.json contains a collection of attack patterns used by scripts/prompt_redteam.py to identify weaknesses in instruction precedence and context boundaries.
[SAFE]: No hardcoded credentials, unauthorized data exfiltration, or obfuscated malicious code were detected. The skill follows security best practices by providing machine-readable outputs and policy-based gating for CI/CD environments.

security-suite