skills/builderio/skills/visual-plan/Gen Agent Trust Hub

visual-plan

Pass

Audited by Gen Agent Trust Hub on Jul 19, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the @agent-native/core and @agent-native/skills packages, as well as the hosted application at plan.agent-native.com. These are verified vendor resources belonging to the BuilderIO/Agent-Native ecosystem used for plan rendering and collaboration.
  • [DATA_EXFILTRATION]: While the skill can publish data to a hosted service, it features a comprehensive 'Local-Files Privacy Mode'. This mode allows users to keep all plan data within their local filesystem and repository, effectively mitigating the risk of unauthorized data exposure for sensitive projects.
  • [COMMAND_EXECUTION]: The skill utilizes shell commands via npx for local plan verification, checking, and serving. These operations are scoped to the local plan directory and are necessary for the interactive preview functionality.
  • [PROMPT_INJECTION]: The skill includes an ingestion surface for existing text plans (planText). The instructions provide strong guidance for the agent to ground the resulting document in real repository files and symbols, which acts as a technical constraint against potential indirect injection from the source material.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 19, 2026, 05:41 AM
Security Audit — agent-trust-hub — visual-plan