dialogue-gym
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes and analyzes user-provided dialogue samples, which presents a surface for indirect prompt injection attacks.
- Ingestion points: Users are encouraged to paste dialogue exchanges for diagnosis and critique in the Intake and Drill sections of SKILL.md.
- Boundary markers: The instructions do not specify the use of delimiters or provide the agent with instructions to disregard malicious prompts embedded within user text samples.
- Capability inventory: The skill has access to external character voice data via an MCP connector, though the documentation indicates this access is read-only.
- Sanitization: No explicit logic or instructions for the agent to sanitize or filter user-provided content are defined.
Audit Metadata