revenuecat-paywall-design
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to search and ingest content from the RevenueCat blog and documentation to provide design advice and perform paywall audits. This creates a potential surface for indirect prompt injection if the external content contains malicious instructions.
- Ingestion points: RevenueCat blog and documentation pages (SKILL.md)
- Boundary markers: Absent; there are no instructions to treat search results as untrusted
- Capability inventory: The skill has write access to paywall configurations via
create-paywall-ai,edit-paywall-ai, andduplicate-paywall(SKILL.md) - Sanitization: Absent; the skill does not specify validation for the gathered design ideas
- [EXTERNAL_DOWNLOADS]: The skill directs the agent to fetch external markdown files from the vendor's web infrastructure to stay updated on best practices.
- Evidence: "Search the RevenueCat blog and docs for paywall design and best practices. Markdown docs are at the page URL with .md appended." (SKILL.md)
- Context: The downloads target the official RevenueCat domains for legitimate educational content.
Audit Metadata