The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a Bash script raw_http_vuln_verify.sh that uses curl to replay and verify HTTP-based vulnerabilities like open redirects and reflected XSS. This is a core function for the skill's stated purpose of offensive security validation.
[DATA_EXFILTRATION]: The skill provides detailed workflows in references/secrets-exposure-workflows.md for discovering and validating the exposure of sensitive files and credentials (e.g., .env files, AWS keys) within authorized targets.
[EXTERNAL_DOWNLOADS]: The documentation in references/pentesting-shannon.md references the use of npx @keygraph/shannon, which facilitates the download and execution of an external autonomous penetration testing framework.
[PROMPT_INJECTION]: The skill instructions in SKILL.md and related reference files use strong instructional language to define the agent's behavior, authorization requirements, and operational scope during pentesting tasks.

pentest