prompt-engineering-pro
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided task descriptions to generate complex prompt structures, which is an inherent surface for indirect prompt injection.
- Ingestion points: User requests for prompt design and optimization described in SKILL.md.
- Boundary markers: While the skill emphasizes system prompts as a 'contract', it lacks explicit instructions for using delimiters or boundary markers when handling potentially malicious user-provided data during the prompt generation process.
- Capability inventory: The skill does not include any executable scripts, tools, or network capabilities, significantly limiting the impact of any potential injection.
- Sanitization: No specific input validation or output sanitization logic is prescribed for the agent to follow.
Audit Metadata