case-study-writing
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill employs the
infshCLI for authentication and to execute specialized assistant tools. - [EXTERNAL_DOWNLOADS]: The documentation references the installation of additional skills from the official and trusted
inference-shGitHub organization. - [REMOTE_CODE_EXECUTION]: The skill uses a Python executor to generate visualizations via the
matplotliblibrary. While the provided code snippet is safe and intended for data visualization, the executor allows for the dynamic processing of code. - [PROMPT_INJECTION]: The skill ingests data from external search providers (Tavily, Exa) into the agent's context. This creates a surface for indirect prompt injection, as untrusted data from the web is processed by an agent that has access to code execution capabilities. The skill does not define specific boundary markers or content sanitization procedures.
Audit Metadata