perfectcode-zen-plan

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes git status and git log to gather codebase context. This is a standard and safe operation for a development-focused planning tool.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted user requirements and external codebase content during the planning phase.
  • Ingestion points: User feature requests and codebase context gathered via git and the @explore agent.
  • Boundary markers: The workflow does not explicitly define delimiters or instructions to ignore embedded malicious commands within the analyzed content.
  • Capability inventory: The skill performs file writes to the .opencode/plans/ directory and executes shell commands for git context.
  • Sanitization: There is no evidence of sanitization or filtering for the data ingested from the user or the codebase before it is synthesized into the final plan.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 10:22 PM
Security Audit — agent-trust-hub — perfectcode-zen-plan