ui-skills-root

Warn

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to fetch the ui-skills package from the NPM registry during execution.
  • [COMMAND_EXECUTION]: The skill requires the agent to run several shell commands including npx ui-skills start, npx ui-skills categories, npx ui-skills list, and npx ui-skills get <slug>.
  • [REMOTE_CODE_EXECUTION]: By instructing the agent to run an unverified community package through npx, the skill triggers remote code execution. The security posture of the host depends on the integrity of the third-party ui-skills package on NPM.
  • [PROMPT_INJECTION]: The skill processes instructions from the CLI tool's output to guide implementation, creating a surface for indirect prompt injection.
  • Ingestion points: Output from the npx ui-skills get <slug> command (SKILL.md).
  • Boundary markers: No delimiters or warnings are used to prevent the agent from following instructions embedded within the tool's output.
  • Capability inventory: The agent is authorized to 'implement using that context', allowing for code generation and subsequent command execution based on untrusted data.
  • Sanitization: There is no evidence of validation or sanitization applied to the CLI output before it is ingested into the agent context.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 5, 2026, 06:48 AM
Security Audit — agent-trust-hub — ui-skills-root