paywall-upgrade-cro
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides frameworks and best practices for in-app paywall design and optimization. The content is purely instructional and does not contain any suspicious commands or patterns.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by instructing the agent to read external context files.
- Ingestion points:
.agents/product-marketing-context.md(referenced in SKILL.md) - Boundary markers: None identified; the agent is simply told to read the context if it exists.
- Capability inventory: No risky capabilities (subprocess, eval, network, file writes) are present in the skill instructions or scripts.
- Sanitization: No input validation or sanitization is performed on the ingested context.
- Conclusion: Due to the total absence of exploitable capabilities, this injection surface is considered safe.
Audit Metadata