senior-prompt-engineer
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides comprehensive informational references and utility scripts for prompt optimization and evaluation. It relies on platform-native features and user-supplied configuration for sensitive calculations like API costs, explicitly instructing the agent not to trust cached price tables.
- [INDIRECT_PROMPT_INJECTION]: While the skill's primary function involves ingesting and analyzing user-supplied data (prompts, RAG contexts, and agent configurations), the associated tools are limited to static analysis and statistical metrics. There is no evidence of the content being executed, used to spawn subprocesses, or transmitted over a network. The scripts act as processors for local data files provided via CLI arguments.
- [COMMAND_EXECUTION]: The skill includes Python utility scripts that operate on local files. These scripts are implemented using the Python standard library, do not utilize risky functions like eval() or exec(), and do not pass unsanitized input to shell commands. The scripts perform parsing, regex matching, and mathematical calculations in a controlled environment.
Audit Metadata