analyze
Pass
Audited by Gen Agent Trust Hub on May 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of instructional prompts designed to guide the agent through a logical analysis process. It does not include scripts, executables, or network requests.
- [PROMPT_INJECTION]: The skill's primary function is to ingest and analyze potentially untrusted repository data and user input via the {{ARGUMENTS}} field. To mitigate the risk of indirect prompt injection, it enforces strict boundary markers by requiring a specific output format that separates direct evidence from inferences. The skill's capabilities are limited to read-only operations, and it mandates concrete file references for all material claims, providing a manual verification path for the user.
- [COMMAND_EXECUTION]: The instructions reference platform-specific parallelization tools such as native subagents and coordination via OMX runtime. These references represent standard usage of the agent's coordination environment for complex tasks and do not involve arbitrary or unsafe shell command execution.
- [DATA_EXFILTRATION]: No network activity, unauthorized file reading of sensitive system paths, or credential harvesting patterns were detected. The analysis is restricted to the context of the user-provided repository.
Audit Metadata