// MCP TOOL REFERENCE

FISHCLAW MCP TOOLS

12 tools from the FishClaw MCP MCP Server, categorised by risk level.

// ALL 12 FISHCLAW MCP TOOLS

READ 3 tools

Read get_page_content 读取当前浏览器页面的可见文字内容（最多 3000 字符），供分析页面状态或提取信息。 Read get_selling_items 获取当前账号所有在售商品列表（标题、价格、链接）。 Read search_market 在闲鱼搜索指定关键词，采集结果列表（标题、价格、链接），用于竞品调研和定价参考。

WRITE 3 tools

Write draft_item 在闲鱼发布页面填写商品草稿（图片、描述、分类、价格），填写完成后自动截图。 Write generate_product_description 根据技术主题生成闲鱼商品描述文案（自我介绍 + 服务分类 + 技术关键词三段式）。 Write publish_item 点击发布按钮完成商品发布。

DESTRUCTIVE 1 tools

Destructive manage_item 对指定商品执行下架或删除操作，内部自动跳转商品详情页并处理确认弹窗。

EXECUTE 5 tools

Execute restart_browser 重新初始化 Playwright 浏览器进程，无需重启 MCP 服务器。 Execute generate_image 根据文本提示词调用阿里云 DashScope 生成图像，图像缓存到本地后返回路径。 Execute generate_image_prompt 根据技术主题生成适合图像生成模型的英文提示词（赛博朋克/科技感风格）。 Execute login 检查闲鱼登录状态。已登录则直接返回；未登录则打开浏览器展示二维码， Execute simulate_farming 模拟真人在闲鱼首页随机浏览（滚动、点击进帖、返回），用于账号养号。

// FAQ

How many tools does the FishClaw MCP MCP server have? +

The FishClaw MCP MCP server exposes 12 tools across 4 categories: Read, Write, Destructive, Execute.

How do I enforce policies on FishClaw MCP tools? +

Route the FishClaw MCP server through the PolicyLayer gateway. Define allow, deny, or approval rules per tool in the dashboard; they are enforced on every call before it reaches the server.

What risk categories do FishClaw MCP tools fall into? +

FishClaw MCP tools are categorised as Read (3), Write (3), Destructive (1), Execute (5). Each category has a recommended default policy.

Enforce policy on every FishClaw MCP tool call.

Start from FishClaw MCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →

Free to start. No card required.

43,000+ MCP servers and 220,000+ tools scanned and risk-classified.