sikaru-inference
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references the installation of
sikaru-sdkvia Python and Node.js package managers. These libraries are the official tools for accessing the vendor's inference services.- [COMMAND_EXECUTION]: The skill provides examples of using thesikaruCLI utility to fetch model metadata and execute inference tasks.- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of user-provided content through its inference capabilities. While it lacks explicit input delimiters, it relies on standard model safety guardrails. - Ingestion points: User-supplied parameters in CLI and SDK chat calls in
SKILL.md. - Boundary markers: Not explicitly defined in the provided examples.
- Capability inventory: Network requests to inference providers and local CLI execution.
- Sanitization: No explicit validation or escaping logic described in the instructions.
Audit Metadata