smtp-penetration-testing
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill defines a variety of shell commands for service discovery, banner grabbing, and vulnerability exploitation using standard tools such as
nmap,nc,hydra,smtp-user-enum, and the Metasploit Framework. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze output from external SMTP servers (e.g., banners, command responses, and enumeration results), which serves as an entry point for untrusted data.
- Ingestion points: Data enters the context through tool outputs from network probes and service interactions defined in
SKILL.mdandreferences/detailed-guide.md. - Boundary markers: The skill implements a procedural boundary through a "Mandatory confirmation gate" that requires the agent to obtain explicit user authorization and scope confirmation before executing any probing or exploitation commands.
- Capability inventory: The skill exercises capabilities including network communication, local tool execution, and access to local wordlists for brute force and enumeration.
- Sanitization: There is no technical sanitization described for the raw string data received from SMTP servers; safety relies on the user-in-the-loop confirmation process.
Audit Metadata