blog-author
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes untrusted data from web research and user-provided documents.
- Ingestion points: Untrusted data enters the agent context via web research tools and the retrieval of context from user materials (SKILL.md, Stage 1).
- Boundary markers: There are no explicit instructions or delimiters used to separate the research data from the agent's operational instructions, which could allow malicious content to influence behavior.
- Capability inventory: The skill utilizes the agent's ability to write files to the local filesystem (Markdown), perform network searches, and generate images.
- Sanitization: No sanitization, escaping, or validation of external content is performed before the data is used in the drafting process.
Audit Metadata