etymology-and-word-origins-expert
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines a research checklist that directs the agent to ingest data from external sources and use it to update its core guidance modules.
- Ingestion points: In
modules/research-checklist.md, the agent is instructed to use tools to fetch data from Wikipedia and Arxiv. - Boundary markers: The instructions lack markers or specific warnings to ignore embedded instructions within the research data, which increases the risk of the agent following malicious commands found in those sources.
- Capability inventory: The agent is explicitly directed to "Update modules/core-guidance.md" based on the results of the external research, allowing untrusted data to potentially modify the skill's future instructions.
- Sanitization: No validation or sanitization steps are defined for the content retrieved from external sources before it is incorporated into the skill's logic.
Audit Metadata