ag-ui
Warn
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to run
npx create-ag-ui-app, which downloads thecreate-ag-ui-apppackage from the npm registry.- [REMOTE_CODE_EXECUTION]: By usingnpx, the skill immediately executes the downloadedcreate-ag-ui-apppackage, constituting a remote code execution pattern for unverified external code.- [COMMAND_EXECUTION]: The skill requires theBashtool to perform scaffolding, installation, and environment initialization commands such asnpxand package manager operations.- [PROMPT_INJECTION]: Analysis of indirect prompt injection surface: - Ingestion points: The agent architecture involves processing external event streams like
TEXT_MESSAGE_CONTENTandSTATE_DELTA(SKILL.md). - Boundary markers: There are no identified delimiters or warnings to ignore instructions embedded in the event data.
- Capability inventory: The skill has access to powerful tools including
Bash,Write, andEdit(SKILL.md). - Sanitization: The skill lacks explicit sanitization or validation protocols for the external data it processes.
Audit Metadata