impeccable

Pass

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Node.js utility scripts (live.mjs, live-server.mjs, pin.mjs, load-context.mjs, etc.) to manage project context and the live iteration environment. It is authorized to run the impeccable CLI tool via npx. The pin.mjs script modifies harness directories to create subcommand shortcuts.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it ingests project documentation and source code to drive its design generation logic.\n
  • Ingestion points: scripts/load-context.mjs (reading PRODUCT.md and DESIGN.md) and scripts/live-server.mjs (reading source files via the /source endpoint).\n
  • Boundary markers: Absent. The skill does not use delimiters to isolate workspace content from its instructions.\n
  • Capability inventory: The skill can modify project source files and execute CLI commands.\n
  • Sanitization: Absent. Workspace content is used directly as context.\n- [EXTERNAL_DOWNLOADS]: The skill uses npx to run the impeccable package, which downloads the package from the npm registry if not already available.\n- [COMMAND_EXECUTION]: The skill generates and writes code variants directly into source files (HTML, CSS, JSX, TSX). This is the intended behavior but constitutes a significant capability for automated filesystem modification.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 11, 2026, 07:06 PM