Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill operates by invoking the
cyeam mailcommand-line utility. It translates user requests into shell commands to interact with IMAP and SMTP servers. - [DATA_EXFILTRATION]: The skill reads account configurations from
~/.cyeam/mail.json. It mitigates the risk of credential exposure by encouraging the use of environment variables (username_envandpassword_env) instead of hardcoding sensitive information in the configuration file. - [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes untrusted data from external sources.
- Ingestion points: Content of emails retrieved via
cyeam mail readas described inSKILL.md. - Boundary markers: Absent; there are no instructions to use delimiters or ignore embedded instructions within the email text.
- Capability inventory: The agent is capable of listing, reading, and sending emails through the CLI tool, which could be manipulated by adversarial content in an email body.
- Sanitization: The instructions do not specify any validation or sanitization of the email content before it is processed by the agent.
Audit Metadata