data-report
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill constructs a brief based on the user's message, which is then used to guide tools and file-writing actions.
- Ingestion points: The instructions in SKILL.md explicitly state that the rest of the user's message should be treated as the brief.
- Boundary markers: There are no structural delimiters or specific prompts to ignore malicious instructions that might be embedded in the user-supplied brief.
- Capability inventory: The agent is instructed to use file-editing tools and the register_open_design_artifact tool, providing a write-access surface that can be influenced by the untrusted brief.
- Sanitization: No validation, filtering, or sanitization of the user-provided brief is performed before the tool calls occur.
Audit Metadata