skill-creator

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as a meta-utility for skill development and does not contain any prompt injection attacks, obfuscation, or unauthorized data access.
  • [COMMAND_EXECUTION]: The skill employs internal XML tags such as <spawn_subagent> and <dispatch_agents> to manage task execution and performance testing. These are controlled orchestration tools utilized for the skill's primary purpose.
  • [PROMPT_INJECTION]: The skill creates a surface for potential indirect prompt injection by ingesting and processing user-defined test prompts. Ingestion points: User-provided prompts are loaded from evals/evals.json into sub-agent task prompts. Boundary markers: XML tags are used to encapsulate task instructions. Capability inventory: The skill can spawn sub-agents and save evaluation results to local file paths. Sanitization: No specific input sanitization or instruction-filtering mechanisms are mentioned for the ingested test content.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 08:49 AM