html-reports-arch

Pass

Audited by Gen Agent Trust Hub on Jul 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a provided shell script scripts/detect-changes.sh to compare the current repository state against previously generated reports by executing git commands and parsing metadata from .claude/reports/_index.js.
  • [COMMAND_EXECUTION]: The agent is instructed to perform reconnaissance using standard utilities such as git ls-files, find, fd, and grep to identify project entry points, build manifests (e.g., package.json, Cargo.toml, go.mod), and documentation.
  • [COMMAND_EXECUTION]: The skill attempts to leverage existing environment-specific tools for metric and dependency extraction, such as tokei, madge, pyright, or dotnet, to populate the architectural data models.
  • [EXTERNAL_DOWNLOADS]: The generated HTML reports are designed to load visualization libraries (Cytoscape.js, ECharts, and Mermaid) from well-known content delivery networks (CDNs) for interactive diagram rendering.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 5, 2026, 01:15 PM
Security Audit — agent-trust-hub — html-reports-arch