company-research-report
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill specifies shell commands in
references/sankey_guide.mdfor environmental setup, including package installation usingpip installand font cache updates withfc-cache. - [PRIVILEGE_ESCALATION]: The skill instructs the agent to use
apt-get installinreferences/sankey_guide.mdto add system-level font support (fonts-noto-cjk), which requires administrative (root) permissions. - [DYNAMIC_EXECUTION]: The skill uses Python scripts to generate visual report components, involving runtime script creation and file system operations to save PNG, HTML, and DOCX outputs.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from web searches (Phase 0) and interpolates it into report text and visualization scripts.
- Ingestion points: External web search results defined in the SKILL.md workflow.
- Boundary markers: The skill does not define specific delimiters or instructions to ignore malicious content within the external data.
- Capability inventory: The agent has permissions for code execution, file writing, and network access via search tools.
- Sanitization: There is no evidence of input validation or escaping for the data retrieved from external sources before it is used in calculations or visualization logic.
Audit Metadata