plan-before-code

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: Analysis of the skill content and configuration found no evidence of malicious patterns, credential harvesting, or unauthorized data access.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from project documentation (e.g., AGENTS.md, README) and user task definitions as seen in SKILL.md. While it lacks explicit sanitization or boundary markers, the skill is limited to information gathering and writing local implementation plans, representing standard operational behavior with no significant security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 11:52 PM
Security Audit — agent-trust-hub — plan-before-code