himalaya-mail
Pass
Audited by Gen Agent Trust Hub on Mar 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
himalayaCLI tool to perform email operations such as listing, reading, and sending messages. - [DATA_EXFILTRATION]: Accesses the email configuration file located at
~/.wegent-executor/mail/config.toml, which contains sensitive information including email credentials and server configurations. - [PROMPT_INJECTION]: Vulnerable to indirect prompt injection attacks where malicious instructions embedded in incoming emails could influence agent behavior.
- Ingestion points: Processes external, untrusted data by reading email bodies and headers via
himalaya message readandhimalaya envelope listcommands. - Boundary markers: The instructions do not define specific delimiters or boundary markers to separate email content from the agent's internal logic.
- Capability inventory: The agent has the capability to send emails, move messages between folders, and delete content using the
himalayaCLI. - Sanitization: No sanitization or filtering logic is specified for the data ingested from email messages.
Audit Metadata