cli
Pass
Audited by Gen Agent Trust Hub on Apr 8, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The documentation provides instructions for executing the CLI using
bunandlhfor development and production tasks. - [CREDENTIALS_UNSAFE]: References the storage of encrypted authentication tokens in
~/.lobehub/credentials.jsonas part of the documented CLI architecture. - [EXTERNAL_DOWNLOADS]: Describes functional features of the CLI for downloading generated assets and installing skills from GitHub or remote URLs.
- [DATA_EXFILTRATION]: Documents file management features including uploads to S3 via presigned URLs and local file system access for configuration storage.
- [PROMPT_INJECTION]: The CLI's generation commands process user-supplied prompts, presenting a standard attack surface for indirect prompt injection common to AI-integrated tools.
Audit Metadata