ian-xiaohei-illustrations

Pass

Audited by Gen Agent Trust Hub on Jul 4, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external, untrusted data. The 'Workflow' section in SKILL.md instructs the agent to read user-provided articles, links, and Notion pages. There are no explicit delimiters or instructions provided to the agent to ignore or isolate embedded commands within the processed text. The skill has the capability to invoke the image_gen tool on supported platforms and perform local file writes, which increases the potential impact of an injection. No sanitization is applied to the ingested text before it influences the conceptual output.
  • [COMMAND_EXECUTION]: The skill performs file system operations to organize and save generated image assets. In SKILL.md, the 'Save the Deliverable' section defines a process for saving PNG files to workspace directories. It incorporates defensive slugging rules, such as ASCII conversion and hyphen replacement, to sanitize directory and filenames, which effectively mitigates common path traversal risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 4, 2026, 10:14 PM
Security Audit — agent-trust-hub — ian-xiaohei-illustrations