arrow-typed-errors
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- PROMPT_INJECTION (SAFE): No attempts to override agent instructions or bypass safety filters were detected. The language is purely instructional and technical.
- DATA_EXFILTRATION (SAFE): No access to sensitive files or unauthorized network calls. The external links provided are to official documentation for the Arrow library (arrow-kt.io).
- EXTERNAL_DOWNLOADS (SAFE): No suspicious package installations or remote script executions. The references to external libraries (Quiver, Pedestal State) are conceptual mentions in documentation.
- COMMAND_EXECUTION (SAFE): No shell commands, privilege escalation, or persistence mechanisms are present. The skill focuses on Kotlin code structure and error modeling.
- OBFUSCATION (SAFE): No encoded content or hidden characters were found.
- INDIRECT_PROMPT_INJECTION (SAFE): The skill provides coding patterns and does not process untrusted external data or inputs that could influence the agent's behavior maliciously.
Audit Metadata