discovery

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security risks were identified during the analysis of this skill.
  • [COMMAND_EXECUTION]: The skill does not contain any shell commands or subprocess execution logic. It is purely instructional for guiding conversational flow.
  • [EXTERNAL_DOWNLOADS]: No external scripts, packages, or remote resources are referenced or downloaded.
  • [DATA_EXFILTRATION]: The skill does not perform any network operations. It specifies writing an artifact to a local project-specific directory (.planning/harness/), which is a standard workflow practice.
  • [PROMPT_INJECTION]: The instructions are focused on legitimate discovery processes (WHO, WHAT, WHEN, WHY) and do not contain attempts to override safety filters or bypass system constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 08:40 AM