skills/xyaz1313/xyskill/xy-close/Gen Agent Trust Hub

xy-close

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script (scripts/atoms-search.py) to search through a database of sales principles (referred to as "atoms").
  • [EXTERNAL_DOWNLOADS]: The search script included with the skill is configured to connect to https://api.xyskill.xyz to perform remote searches. This domain belongs to the skill's vendor ecosystem and is used to extend the knowledge base beyond local files.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided sales conversation transcripts and "话术" (sales scripts) to provide diagnosis.
  • Ingestion points: User-provided dialogue snippets and sales scenarios submitted during the Phase 1 and Phase 2 diagnostic flows (e.g., "帮我看看成交话术").
  • Boundary markers: Absent. The skill does not use specific delimiters or instructions to prevent the agent from following directives potentially hidden within user-provided sales text.
  • Capability inventory: The agent has the capability to execute the atoms-search.py script via subprocess.
  • Sanitization: The Python search script performs keyword extraction and JSON parsing; it does not execute the user's input as code.
  • [DYNAMIC_EXECUTION]: The skill relies on runtime execution of the Python-based search tool to dynamically retrieve diagnostic evidence based on user input.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 03:29 AM