gate-info-liveroomlocation

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill employs strong directives to override agent behavior and prioritize specific instructions, including a link to external runtime rules (e.g., 'STOP — You MUST read and strictly follow').
  • [EXTERNAL_DOWNLOADS]: The skill fetches its primary runtime logic and safety rules from an external Markdown file hosted on the vendor's GitHub repository ('gate-runtime-rules.md').
  • [COMMAND_EXECUTION]: The documentation references external installer tools for various platforms (e.g., 'gate-mcp-cursor-installer', 'gate-mcp-claude-installer') to set up the required MCP server environment.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it processes and displays titles and names directly from an external API response.
  • Ingestion points: Data enters the context via the 'data.list' field from the 'GET /live/gate_ai/tag_coin_live_replay' endpoint.
  • Boundary markers: The skill does not define specific delimiters or instructions to treat the API-provided titles as untrusted data.
  • Capability inventory: The skill performs network GET requests to the vendor's API.
  • Sanitization: There is no evidence of sanitization or filtering for the titles (live.name or video.title) before they are displayed to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 04:07 AM