bro
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a set of pure natural language instructions with no executable code, tool definitions, or network capabilities. It follows the intended purpose of reformatting text output.
- [INDIRECT_PROMPT_INJECTION]: The skill processes the conversation history (the last message) to restate it. While it ingests untrusted data, the skill has no execution capabilities (as indicated by the disable-model-invocation flag), which prevents any malicious payload in the input from performing unauthorized actions.
- Ingestion points: Processes the previous message in the conversation context.
- Boundary markers: Not explicitly defined.
- Capability inventory: No tools or command execution capabilities are enabled or requested.
- Sanitization: None, as the output is intended to be natural language text only.
Audit Metadata