canva-visual-assets
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends several third-party MCP servers for image search and processing, specifically
imagebank-mcp,stocky, and@ahmaddioxide/mcp-image-resolver. - [COMMAND_EXECUTION]: The documentation includes instructions for users to run commands such as
npx -y imagebank-mcpanduvx stocky-mcpto install and configure external image bank tools. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process arbitrary user input to generate AI prompts and search queries, which presents a standard ingestion surface for indirect prompt injection.
- Ingestion points: User-provided descriptions for images, design briefs, and search queries (documented in
SKILL.md). - Boundary markers: No specific delimiters or instructions to ignore embedded commands are used when interpolating user input into tool parameters.
- Capability inventory: Access to Canva design tools (
generate-image,create-design,export-design) and stock image tools (search_images,download_image) which can write files to the project directory. - Sanitization: There is no evidence of input validation or sanitization before passing user-supplied text to image generation or search APIs.
Audit Metadata