molykit

Pass

Audited by Gen Agent Trust Hub on Apr 19, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill defines structures for processing external data from LLM providers and user inputs, which is an inherent surface for indirect prompt injection in chat applications.
  • Ingestion points: Message data processed in BotClient streams and the PromptInput widget within SKILL.md.
  • Boundary markers: The provided code snippets do not illustrate explicit boundary markers or instruction-guarding techniques to separate developer instructions from processed data.
  • Capability inventory: The toolkit provides capabilities for asynchronous task spawning and file-based asset loading for UI components.
  • Sanitization: No specific input validation or sanitization logic is shown in the example code for the provided message protocol.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 19, 2026, 07:04 PM