unnarrate
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted content such as UI copy, documentation, code comments, and pull request descriptions. This processing task creates an attack surface for indirect prompt injection.\n
- Ingestion points: The skill ingests various external text formats through instructions in SKILL.md and reference/terminology.md.\n
- Boundary markers: There are no explicit markers or instructions defined to prevent the agent from following commands embedded within the processed text.\n
- Capability inventory: The skill is instructional and does not utilize tools for network access, file writing, or command execution.\n
- Sanitization: No sanitization or validation of the input text is required or performed by the skill.\n- [SAFE]: The skill references a wide range of external academic papers, technical documentation, and community repositories from well-known sources (e.g., Oxford University Press, Wikipedia, arXiv, and the Python Software Foundation). These references provide technical context and are documented neutrally.
Audit Metadata