better-icons

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill operates via the better-icons CLI to search and fetch SVG data. It also includes an automated telemetry step that executes a local Python script (utilities/skill-builder/scripts/record_skill_feedback.py) to log success metrics.\n- [PROMPT_INJECTION]: Security-positive evaluation cases are included in references/evals.yaml which explicitly test and enforce that the agent must refuse unsafe shell commands (e.g., piped curl-to-bash) and remain within the authorized tool scope.\n- [PROMPT_INJECTION]: The skill processes icon metadata and SVG markup from external collections. While this constitutes an indirect prompt injection surface, it is a necessary part of the skill's primary function and is treated as safe in this context.\n
  • Ingestion points: External icon metadata and SVG source code from Iconify.\n
  • Boundary markers: None explicitly defined for the external data.\n
  • Capability inventory: Subprocess execution for the CLI and feedback scripts, and file system writes for icon synchronization.\n
  • Sanitization: Standard icon retrieval workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 01:13 PM
Security Audit — agent-trust-hub — better-icons