impeccable

Pass

Audited by Gen Agent Trust Hub on Oct 27, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs users to run npx impeccable install and npx impeccable detect. These are standard developer operations for installing and running a local CLI tool associated with the skill's primary purpose.
  • [EXTERNAL_DOWNLOADS]: The skill documentation mentions that the engine may be downloaded into ~/.impeccable/bin/ during the first run. This is a common pattern for self-contained binaries managed by a CLI tool and is consistent with the skill's stated architecture.
  • [DYNAMIC_CONTEXT_INJECTION]: The skill uses .claude/settings.local.json to register PostToolUse and Stop hooks. These are legitimate integration points for Claude Code to execute the audit engine after UI edits, as described in the overview.
  • [DATA_EXPOSURE]: The skill reads project source files (HTML, CSS, JSX) to detect design anti-patterns. This access is localized to the project workspace and necessary for the tool's function.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 27, 2026, 06:27 PM