metforge-figure
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of plotting scripts and HPC job submissions using Slurm (e.g.,
sbatch). The instructions emphasize reusing validated project configurations for accounts and partitions to ensure session safety. - [DATA_EXPOSURE]: The skill ingests scientific data formats (NetCDF, xarray, tabular) and reads existing user code to generate or audit figures. Access is scoped to data preparation and rendering within the user's project environment.
- [PROMPT_INJECTION]: The skill ingests untrusted data and code (Ingestion points: NetCDF data, user scripts). While boundary markers and sanitization are absent, the skill's capabilities are focused on scientific visualization and it includes explicit rules against credential logging, minimizing the risk of multi-step exploitation.
Audit Metadata