sleek-design-mobile-apps
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches design project metadata, API specifications, and screenshots from the vendor's domain
sleek.design. It also downloads icon SVGs from the well-known serviceapi.iconify.designto support UI implementation tasks. - [COMMAND_EXECUTION]: Instructs the agent to use
curlfor API interactions, specifically recommending that response bodies be saved directly to files. This is intended to efficiently manage large data payloads and prevent zsh shell expansion from corrupting JSON content. - [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to parse and implement HTML code retrieved from the Sleek API, creating a vector where malicious instructions could be embedded in the design data.
- Ingestion points: Component HTML code fetched from the
/api/v1/projects/:id/components/:componentIdendpoint, as described in the 'Implementing Designs' section. - Boundary markers: The instructions do not define specific delimiters or provide 'ignore embedded instructions' warnings for the agent to use while processing the external HTML code.
- Capability inventory: The agent possesses capabilities for shell command execution (
curl), file system writes, and network operations. - Sanitization: The skill does not specify any sanitization, filtering, or validation steps for the HTML content before the agent extracts design tokens such as font links and icon names.
Audit Metadata