editorial-portrait-composition-by-1l42ctj
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read the local
references/cases.mdfile and use tools to inspect external media and source links from third-party sites. - Ingestion points: The agent is directed to visit external URLs from
goodcase.aianddouyin.comto analyze images and text summaries. - Boundary markers: No specific security delimiters or instructions are provided to treat the external site content as untrusted.
- Capability inventory: The agent has the ability to browse the web, inspect images, and generate complex text artifacts (prompts).
- Sanitization: There is no evidence of sanitization or filtering applied to the content retrieved from external sources before it influences the agent's output.
- [EXTERNAL_DOWNLOADS]: The
references/cases.mdfile contains numerous links to external platforms. - Evidence: References to
goodcase.ai,media.goodcase.ai, anddouyin.comare used as primary sources for the skill's creative process.
Audit Metadata