muapi-url-to-design
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input via the
url,page_type, andredesign_stylefields and interpolates them directly into prompts for image generation models.\n - Ingestion points:
url,page_type, andredesign_styleinputs defined inSKILL.md.\n - Boundary markers: None present; inputs are concatenated into raw prompt strings (e.g.,
Professional UI/UX redesign of a {{page_type}} for the website at {{url}}).\n - Capability inventory: The skill uses
muapiCLI commands andcurlfor network requests to external API endpoints.\n - Sanitization: None detected; the skill does not include instructions to validate or escape these inputs before processing.\n- [COMMAND_EXECUTION]: The skill instructions include
curlcommands to interact with a third-party API (api.muapi.ai). These commands transmit the user'sMUAPI_API_KEYin the request header. While this is part of the intended functionality, it represents an automated network operation involving sensitive credentials.
Audit Metadata