blog-geo
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external content (blog posts), creating a vulnerability surface for indirect prompt injection attacks.
- Ingestion points: Step 1 of the audit process in
SKILL.mdinstructs the agent to extract full content text, headings, paragraphs, schema markup, and other metadata from external blog posts. - Boundary markers: The instructions lack specific boundary markers (e.g., XML tags or delimiters) or explicit warnings for the agent to ignore any natural language instructions embedded within the blog post content.
- Capability inventory: The skill uses the ingested data to perform scoring (Step 9) and generate a detailed report (Step 10). A malicious blog post could contain hidden text designed to override these instructions to force a high score or include deceptive content in the generated report.
- Sanitization: There are no instructions provided to sanitize, validate, or escape the content extracted from the blog posts before the agent processes it for analysis.
Audit Metadata