backtest
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
bashtool to execute Python scripts for calculating financial metrics and historical strategy performance. - [DYNAMIC_EXECUTION]: The skill creates and executes standalone Python scripts using
write_fileandbashto handle customized backtesting logic. - [INDIRECT_PROMPT_INJECTION]: The skill features an attack surface for indirect prompt injection when reading untrusted data. 1. Ingestion points: Reads existing Python files from the
scripts/directory and user-provided strategy descriptions. 2. Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious commands embedded in the files it reads. 3. Capability inventory: The skill utilizeswrite_file,bash, andread_filetools, which grant it the ability to modify the workspace and execute arbitrary code. 4. Sanitization: There is no mentioned validation or sanitization process for existing scripts before the agent reviews and incorporates their logic.
Audit Metadata