canva-policy-guardrails
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references official Canva developer documentation for security guidelines, OAuth scopes, and API versioning. These sources are established platforms for Canva Connect integrations.
- [INDIRECT_PROMPT_INJECTION]: The skill processes repository files and policy profiles to verify security invariants. Ingestion points: Ingests content from the specified repository path and policy profile provided as arguments to the skill. Boundary markers: Relies on structured classification of controls to distinguish immutable security invariants from versioned provider data. Capability inventory: Employs Read, Grep, Write, and Edit tools to scan repository content and apply policy configurations. Sanitization: Includes explicit instructions to redact sensitive evidence and return only stable reason codes.
Audit Metadata