ui-typography

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes a shell command example using sed for bulk replacement of straight quotes with curly quotes in .tsx files. This identifies a capability for file system modification, though it is applied here for legitimate formatting tasks.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external UI code and components for auditing purposes, which creates an attack surface. 1. Ingestion points: UI code and components provided for audit or generation as described in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: File modification via sed shell command as described in SKILL.md. 4. Sanitization: Absent.
  • [PROMPT_INJECTION]: The skill contains directives for the agent to auto-apply rules silently and not ask permission. These instructions aim to automate the formatting process without requiring per-action user confirmation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 02:56 PM
Security Audit — agent-trust-hub — ui-typography