writing-plans
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Prompt Injection (LOW): The skill includes hardcoded directives in its output templates (e.g., 'REQUIRED SUB-SKILL: Use superpowers...') intended to override or mandate specific AI agent behaviors.
- Indirect Prompt Injection (LOW): The skill acts as a surface for indirect prompt injection by generating executable tasks from external design data. 1. Ingestion points: User-supplied design and feature descriptions. 2. Boundary markers: Absent from the generated Markdown templates. 3. Capability inventory: Generates Python code snippets and shell commands for 'git' and 'pytest' for subsequent execution by other skills. 4. Sanitization: No logic is present to sanitize or escape external content before it is placed into plan templates.
- Dynamic Execution (LOW): The skill dynamically generates source code and shell scripts based on its internal templates.
Audit Metadata