skills/sunqb/ccsdk/style-transform/Gen Agent Trust Hub

style-transform

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWNO_CODE
Full Analysis
  • [NO_CODE] (SAFE): The provided skill consists entirely of JSON data files (config/ and templates/). No Python, JavaScript, or shell scripts are present, eliminating risks associated with malicious code execution.
  • [PROMPT_INJECTION] (SAFE): Analysis of the text templates and style rules found no evidence of instructions designed to bypass agent safety filters or override system prompts.
  • [DATA_EXFILTRATION] (SAFE): There are no hardcoded secrets, API keys, or sensitive local file paths. No network-capable commands (e.g., curl, fetch) are defined.
  • [REMOTE_CODE_EXECUTION] (SAFE): The skill does not define any external dependencies or download mechanisms for remote content.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The podcast_templates.json file uses placeholders like 【主题】 and 【自定义主题】 for user-provided data. While this creates a surface for indirect prompt injection, the skill's lack of executable capabilities (it only generates text) limits the impact to the generated content itself.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 12:42 PM