blog-schema

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes content from a user-provided file, which serves as a potential vector for indirect prompt injection.
  • Ingestion points: Blog post data is extracted from the file path specified in the argument-hint defined in SKILL.md.
  • Boundary markers: The instructions do not define specific prompt delimiters for the input file, but they include strict requirements to prevent output manipulation.
  • Capability inventory: The skill's primary function is data extraction and schema formatting; it does not request tools for network exfiltration or shell command execution.
  • Sanitization: The skill includes a dedicated security requirement in Step 9 that mandates the agent use a real JSON encoder and escape literal script sequences (e.g., replacing less-than characters) to ensure the generated JSON-LD is safe for HTML embedding.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 04:57 AM
Security Audit — agent-trust-hub — blog-schema