khazix-writer
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external materials including PDFs, news links, briefs, and voice-to-text transcripts to generate long-form articles.
- Ingestion points: Identified in SKILL.md, which instructs the agent to process various user-supplied inputs as source material.
- Boundary markers: The instructions lack explicit boundary markers or directions for the AI to ignore embedded instructions within the provided source materials.
- Capability inventory: The skill is limited to content generation and does not exhibit high-risk capabilities such as network exfiltration, file system writes, or command execution.
- Sanitization: There is no evidence of data sanitization or validation logic to filter potentially malicious instructions embedded in the ingested data.
Audit Metadata