lgs-search

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data in the form of card lists (text or JSON) which are then used to drive automated browser interactions and marketplace searches.
  • Ingestion points: Card lists are provided via the --input flag in SKILL.md.
  • Boundary markers: None specified for the input data.
  • Capability inventory: The skill uses Playwright for browser automation, network requests for price sweeping, and filesystem writes for audit sidecars.
  • Sanitization: Not explicitly described; the skill treats card names as search terms for MTG storefronts.
  • [EXTERNAL_DOWNLOADS]: The skill downloads the Chromium browser binary as part of its setup process.
  • Evidence: uv run playwright install chromium in SKILL.md.
  • Context: This is a standard and required step for applications using the Microsoft Playwright framework.
  • [COMMAND_EXECUTION]: The skill requires the execution of several CLI commands for setup and operation.
  • Evidence: uv sync, playwright install, and lgs-search login commands are documented in SKILL.md.
  • Context: These commands are used for dependency management, browser installation, and store authentication.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 03:24 AM
Security Audit — agent-trust-hub — lgs-search