agent-creator
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION] (LOW): The skill possesses a surface for indirect prompt injection where user-supplied goals and expertise descriptions are used to populate agent definitions. 1. Ingestion points: User input for the primary goal and specialization domain in SKILL.md Step 1. 2. Boundary markers: Structured Markdown headings and YAML frontmatter in assets/agent_template.md. 3. Capability inventory: The resulting agents are templated with Bash, Read, Write, and Edit tools, creating a high-impact capability set. 4. Sanitization: The skill does not perform sanitization or validation of the user-provided text before writing it to the .agent/agents/ directory.
Audit Metadata