tui-front-ui

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it instructs the agent to transform user-supplied design descriptions into structured tool operations. This creates a trust boundary issue where malicious instructions in a user request could influence the generated tool calls.
  • Ingestion points: User requests for TUI designs processed via the SKILL.md instructions.
  • Boundary markers: No explicit boundary markers or instructions to sanitize user input are present in the skill definition.
  • Capability inventory: The skill utilizes tools such as batch_design, open_document, get_editor_state, and snapshot_layout.
  • Sanitization: The instructions lack specific validation or sanitization logic for the data interpolated into the PENCIL_BATCH_DESIGN operations.
  • [COMMAND_EXECUTION]: The skill uses a domain-specific language (DSL) to generate operations for a design tool at runtime. This dynamic generation of execution plans (e.g., in the PENCIL_BATCH_DESIGN block) is based on interpreted user input and results in the agent performing automated design tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 04:07 PM
Security Audit — agent-trust-hub — tui-front-ui