gws-gmail-send

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the gws CLI tool to perform Gmail operations, specifically sending messages and managing file attachments.
  • [PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by accepting arbitrary text for email bodies and subjects which may originate from untrusted external data.
  • Ingestion points: Email subject and body flags used in the CLI command.
  • Boundary markers: None present; the skill treats the entire input string as the email content.
  • Capability inventory: The gws tool can read local files via the --attach flag and send data over the network via SMTP/API.
  • Sanitization: No explicit input validation or escaping logic is described in the instruction set.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 05:06 PM
Security Audit — agent-trust-hub — gws-gmail-send