wtf
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to simplify its previous messages, which creates a surface for indirect instructions. If the previous context contains malicious content from an external source, it could influence the simplification output.
- Ingestion points: The skill processes the agent's previous message history as defined in SKILL.md.
- Boundary markers: Absent; there are no delimiters provided to separate the context being processed from the simplification instructions.
- Capability inventory: None; the skill is a text-based utility and does not include any tools, network access, or system command execution capabilities.
- Sanitization: Absent; the skill does not perform validation or filtering on the text being simplified.
Audit Metadata