decision-support
Pass
Audited by Gen Agent Trust Hub on Mar 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is configured to execute a local entry script (run.py) using the Bash tool. This is a powerful capability that allows the agent to run arbitrary Python logic within the local environment.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface due to its data ingestion model.
- Ingestion points: The skill allows loading external market or account data via the --csv argument, which accepts arbitrary local file paths.
- Boundary markers: The skill instructions do not specify any boundary markers or instructions to the agent to disregard commands or formatting embedded within the ingested data.
- Capability inventory: The skill has access to Bash, Read, and Glob tools, which could be misused if an attacker successfully injects instructions into the data files processed by the agent.
- Sanitization: There is no evidence of content validation or sanitization for the data ingested from external CSV files.
Audit Metadata