gtd-coach
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- [General] (SAFE): The skill only provides text-based templates for session reviews (daily, weekly, etc.) and goal tracking.
- [Category 1: Prompt Injection] (SAFE): No instructions attempt to override agent safety filters or system prompts.
- [Category 2: Data Exposure] (SAFE): The skill does not access sensitive local file paths or make network requests.
- [Category 4 & 10: Code Execution] (SAFE): There are no code execution patterns, runtime compilations, or external downloads. All 'logic' described is natural language instruction for the agent's behavior.
- [Category 8: Indirect Prompt Injection] (LOW): While the templates are designed to process user-provided task descriptions, the lack of executable tools or network capabilities limits the risk to minor reasoning influence within the local session.
Audit Metadata