Menu
构建视觉调用器并可执行单次图片描述
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
输出白名单群信息结构
聚合新闻/热搜/段子/历史/汇率等信息并返回可读摘要
查询指定城市当前天气或未来 1-16 天天气预报，并返回简洁天气文本
模型路由参数规范化与 data URL 解析辅助
按用户明确要求生成图片并发送给聊天。
先由 LLM 理解资源需求并生成搜索计划，再执行搜索并整合结果
| name | vision_caller |
| description | 构建视觉调用器并可执行单次图片描述 |
| entrypoint | main.py |
| parameters | {"type":"object","properties":{"action":{"type":"string","description":"build 或 describe"},"api_type":{"type":"string","description":"模型类型，如 openai、gemini_official、anthropic"},"api_key":{"type":"string","description":"视觉模型 API Key"},"base_url":{"type":"string","description":"API 基础地址"},"model":{"type":"string","description":"模型名称"},"prompt":{"type":"string","description":"describe 时传入的提示词"},"image_url":{"type":"string","description":"describe 时传入的图片 URL 或 data URL"}},"required":[]} |
用于视觉能力自检与单次图片识别验证。
