abaqus-material
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No attempts to override agent behavior or bypass safety guidelines were found. The instructions are strictly limited to the domain of material science and Abaqus scripting.
- [DATA_EXFILTRATION]: No sensitive file paths, hardcoded credentials, or unauthorized network operations were detected in the skill body or metadata.
- [REMOTE_CODE_EXECUTION]: No remote script downloads or executions from untrusted sources were identified. The skill relies on local environment tools.
- [COMMAND_EXECUTION]: The skill utilizes a scoped bash tool
Bash(abaqus:*), which restricts command execution to specific Abaqus-related contexts, limiting the risk of arbitrary command injection. - [OBFUSCATION]: No obfuscated content, encoded strings, hidden characters, or homoglyph attacks were found in the analyzed files.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided material properties and names. While this presents a potential attack surface, the instructions include validation checklists (e.g., verifying Poisson's ratio and density) that encourage safe handling of data. No dangerous interpolation of untrusted data into shell commands was observed.
Audit Metadata