pencil

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides structured documentation for using the Pencil MCP (Model Context Protocol) server to read and modify .pen design files.
  • [DATA_EXPOSURE]: The skill explicitly states that .pen files are encrypted and should only be accessed through the provided MCP tools, preventing accidental data exposure through standard file reading methods.
  • [COMMAND_EXECUTION]: The provided MCP configuration includes a local binary path (/Users/wandl/...) as a reference for the execution environment. This is a standard configuration pattern for MCP servers and does not introduce external security risks.
  • [INDIRECT_PROMPT_INJECTION]: While the skill ingests data from design files (e.g., via batch_get), it includes boundary instructions requiring users to review change plans before execution, mitigating risks from untrusted design data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 04:13 AM
Security Audit — agent-trust-hub — pencil