structured-data
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and process data from external sources, creating a surface where malicious instructions could influence the agent's behavior.
- Ingestion points: The agent is directed to "Inspect the rendered page" and read content from
.agents/marketing-context.md(SKILL.md), both of which can contain untrusted or third-party input. - Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat embedded commands in the external data as untrusted or to ignore them.
- Capability inventory: The agent uses the ingested data to generate JSON-LD markup, rationales, and provides workflow guidance (structured-data-workflow.md).
- Sanitization: The instructions lack specific routines for sanitizing or filtering the content fetched from rendered pages before it is interpolated into the agent's context.
Audit Metadata