skills/ccch1mneyyy/dsh-go/dsh-go-ui/Gen Agent Trust Hub

dsh-go-ui

Pass

Audited by Gen Agent Trust Hub on Nov 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed to manage a local configuration file (ui.json) for the dsh-go mobile application. It includes a dedicated validation script (check.mjs) that mimics the app's internal parser to ensure the configuration is valid before application.
  • [SAFE]: All UI actions defined in the configuration (such as sending messages, executing slash commands, or opening links) are explicitly designed to require manual user interaction. The skill documentation emphasizes that messages are never sent automatically and links require a confirmation prompt before opening.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for defining UI actions (like pre-filling the chat input) which could be used to store malicious prompts if the agent is influenced by untrusted external data. However, the skill implements a strict validation script (check.mjs) and the application requires the user to manually trigger any action and confirm message delivery, effectively mitigating the risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 19, 2026, 01:08 PM