The Agent Skills Directory

[SAFE]: The skill provides reference documentation for legitimate security tools and incorporates extensive safety warnings and operational constraints intended to ensure authorized use.
[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for ingesting and processing results from external targets through security scanning tools, which introduces a potential attack surface for indirect prompt injection from compromised or adversarial targets.
Ingestion points: Network enumeration results, identified vulnerabilities, and web findings mentioned in the Core Workflow of SKILL.md.
Boundary markers: The skill requires a mandatory confirmation gate where the user must state the target, confirm authorization, and approve exact commands before execution.
Capability inventory: Includes instructions for powerful exploitation tools such as Metasploit (msfconsole), database injection tools (sqlmap), and credential crackers (hydra, john).
Sanitization: There are no explicit instructions for the agent to sanitize or filter the content returned by these external tools before processing it in the conversation context.

pentest-commands