plan-canvas

Pass

Audited by Gen Agent Trust Hub on Jul 8, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the ecc-plan-canvas CLI tool to open artifacts, wait for user feedback, and manage the review session.
  • [SAFE]: The skill operates a local loopback server (127.0.0.1:4517) and stores its internal state in a dedicated local directory (~/.claude/plan-canvas/), which is standard practice for local developer tools to maintain session data.
  • [SAFE]: External dependencies for visual rendering, such as Mermaid diagrams, are loaded from a pinned CDN in the browser, which is a routine and legitimate method for providing web-based visualizations.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 8, 2026, 10:11 PM
Security Audit — agent-trust-hub — plan-canvas