ui-ux-pro-max
Fail
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for the agent to download and install the full version of the tool from an unverified third-party GitHub repository (
nextlevelbuilder/ui-ux-pro-max-skill) using thenpx skills addand/plugincommands. This bypasses standard package verification for the content being installed. - [REMOTE_CODE_EXECUTION]: The skill requires the execution of a bundled Python script (
scripts/search.py) to query its internal database. Running scripts from unverified external sources allows for potential arbitrary code execution on the host system. - [COMMAND_EXECUTION]: The installation and detection workflow involves executing shell commands provided in the
SKILL.mdfile, including environment checks (grep,echo) and package management operations (npx). - [INDIRECT_PROMPT_INJECTION]: The skill demonstrates a potential surface for indirect prompt injection.
- Ingestion points: Data is read from a "local design-intelligence database" (referenced in SKILL.md).
- Boundary markers: None are documented to separate database content from agent instructions.
- Capability inventory: The skill executes Python subprocesses (
search.py) and has the capability to persist design systems to files (referenced in SKILL.md). - Sanitization: No sanitization or validation of the database content is documented, meaning malicious instructions embedded in the design database could influence agent behavior.
Recommendations
- AI detected serious security threats
Audit Metadata