cfd-simulation-setup
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTIONDATA_EXFILTRATIONPRIVILEGE_ESCALATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes plain-language user requests to configure CFD simulations. This ingestion of untrusted data creates a vulnerability surface for indirect prompt injection, particularly as the agent has the capability to generate and execute shell scripts based on these inputs.\n- [COMMAND_EXECUTION]: The skill generates and executes bash scripts (Allrun.agent) to manage the simulation workflow. While this is the intended purpose, it represents a significant capability that requires strict parameter validation.\n- [DYNAMIC_EXECUTION]: The scripts/build_case.py script dynamically constructs shell scripts from numeric templates and user-controlled parameters, representing a potential execution risk if parameters are not properly sanitized.\n- [DATA_EXFILTRATION]: The build_case.py script utilizes a user-controlled --template argument to copy directory structures using shutil.copytree. This could potentially be exploited to copy sensitive local directories into the public simulation workspace if path validation is bypassed.\n- [PRIVILEGE_ESCALATION]: The skill automatically modifies file permissions using chmod +x (0o755) on dynamically generated shell scripts to enable execution.
Audit Metadata