blog-post

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection Surface: The skill is designed to ingest data from a research sub-agent via a generated file (research/[slug].md). This creates an attack surface where malicious instructions embedded in research sources could potentially influence the agent's subsequent writing tasks.
  • Ingestion points: The agent reads the findings from research/[slug].md before drafting the post.
  • Capability inventory: The skill allows writing to the file system (within the blogs/ directory) and invoking image generation tools.
  • Sanitization: The skill does not explicitly instruct the agent to use boundary markers or to ignore instructions found within the research data, which is a common security consideration when handling external inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 07:29 PM
Security Audit — agent-trust-hub — blog-post