mailtester

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to compile a local binary using go build and subsequently execute it. This involves running code with the user's local permissions.
  • [EXTERNAL_DOWNLOADS]: The skill references dependencies from the emersion organization's GitHub repositories (github.com/emersion/go-smtp and github.com/emersion/go-imap/v2) which are fetched during the build process. These are well-known libraries in the Go community.
  • [DATA_EXFILTRATION]: The tool is designed to handle sensitive credentials via the -user and -pass flags. These credentials are transmitted to the specified remote host during authentication tests.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it ingests and processes data from external SMTP/IMAP servers and presents that data to the agent.
  • Ingestion points: Terminal output from the ./mailtester binary, which includes server greetings, capability strings, and mailbox metadata fetched from remote hosts.
  • Boundary markers: None present; the agent processes the raw output of the diagnostic tool.
  • Capability inventory: The skill can execute a local binary and perform network operations to arbitrary hosts and ports.
  • Sanitization: There is no evidence of sanitization or filtering of the content received from remote servers before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 10:22 PM
Security Audit — agent-trust-hub — mailtester