htmlcss-to-image

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Installs the @membranehq/cli package from the npm registry. This is the official command-line interface provided by the vendor to interact with their platform services.
  • [COMMAND_EXECUTION]: Executes shell commands via the membrane CLI to perform actions such as user authentication, connector searching, and running image generation tasks. These operations are within the scope of the skill's primary purpose.
  • [CREDENTIALS_UNSAFE]: Implements secure authentication by leveraging the membrane login flow. This design pattern ensures that API keys and tokens are managed server-side by the platform rather than being stored as plaintext in the skill configuration.
  • [PROMPT_INJECTION]: The skill processes external data, such as public URLs and user-provided HTML/CSS, which represents a surface for indirect prompt injection.
  • Ingestion points: HTML input in create-image-from-html and URL input in create-image-from-url.
  • Boundary markers: No explicit delimiters are defined in the command templates for external content.
  • Capability inventory: Shell command execution via the membrane CLI.
  • Sanitization: None observed within the skill instructions; processing relies on the downstream service's rendering environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 11:47 PM