bx-oshi
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface by ingesting data from the host operating system, such as process names (
os.getProcesses()), disk identifiers, and network interface names. This untrusted content is pulled into the agent's context without explicit boundary markers or sanitization logic in the provided patterns. - [COMMAND_EXECUTION]: The skill documentation guides users to install the
bx-oshimodule using standard package management tools likeinstall-bx-moduleandbox install. These commands are standard for extending the vendor's runtime with the necessary telemetry capabilities.
Audit Metadata