chainlit

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill provides instructions for installing the chainlit package and integrates with other common libraries such as openai, langchain, and llama-index. These are well-known, standard dependencies in the AI ecosystem.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The framework documentation explains how to process untrusted data from user messages and file uploads. While this establishes a vulnerability surface for applications built using the framework, the risk to the agent using this skill is minimal and limited to its reasoning when generating code.
  • Ingestion points: User messages (cl.on_message) and file uploads (cl.AskFileMessage) across multiple reference files.
  • Boundary markers: Not explicitly defined in documentation examples.
  • Capability inventory: The skill is used for architectural guidance and code generation.
  • Sanitization: No explicit sanitization or filtering is demonstrated in the basic examples.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 04:39 AM