himalaya
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download an installation script from the official Himalaya project repository on GitHub.
- [REMOTE_CODE_EXECUTION]: The suggested installation method involves piping a remote script directly to a shell, which executes code from an external source.
- [PROMPT_INJECTION]: The skill is subject to indirect prompt injection as it processes untrusted data from incoming emails.
- Ingestion points: Untrusted data enters the agent context via email bodies and headers when using the list and read commands in SKILL.md.
- Boundary markers: There are no boundary markers or instructions to the agent to disregard potential commands embedded in the email content.
- Capability inventory: The agent has access to the himalaya CLI, which can perform network operations and file system writes (attachment downloads) across all scripts.
- Sanitization: The skill does not perform any sanitization or validation on the content of the emails before they are processed by the agent.
Audit Metadata