metforge-model-diagnose
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill is designed to interact with High-Performance Computing (HPC) environments, directing the agent to use
sbatchand Slurm for job scheduling and model integrations. It also prescribes the use of shell commands for lightweight system and configuration inspections. - [DATA_EXPOSURE]: The instructions require the agent to maintain a processing record, appending job metadata, inputs, and environment details to log files (e.g., in
output/logs/). The skill includes a specific safety guideline to ensure no credentials or secrets are recorded in these logs. - [DATA_INGESTION]: The skill establishes an attack surface for indirect prompt injection by processing external data sources.
- Ingestion points: The agent is instructed to inventory and inspect actual run configurations (namelists), compiler settings, and model outputs (
SKILL.md). - Boundary markers: There are no explicit markers or instructions to ignore embedded natural language directives within the ingested configuration files or model output metadata.
- Capability inventory: The skill possesses the capability to execute shell commands through the system scheduler and perform file-write operations for logging (
SKILL.md). - Sanitization: The instructions do not define specific validation or sanitization routines for the content of the model outputs or namelists before they are processed by the agent.
Audit Metadata