create-agent-skill

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • Indirect Prompt Injection (LOW): This skill acts as a generator for other agent skills based on user-provided arguments. 1. Ingestion points: The $ARGUMENTS variable in SKILL.md accepts arbitrary user text. 2. Boundary markers: There are no delimiters or 'ignore instructions' warnings around the user input. 3. Capability inventory: The skill invokes the create-agent-skills tool to generate new skill definitions. 4. Sanitization: No sanitization or validation of the input is performed before it is used to generate agent instructions.
  • No Code (SAFE): The skill consists only of a markdown file with instructional metadata and does not include any scripts or executable binaries.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:10 PM