# TOOLS.md - Local Notes

Skills define _how_ tools work. This file is for _your_ specifics — the stuff that's unique to your setup.

## What Goes Here

Things like:

- Camera names and locations
- SSH hosts and aliases
- Preferred voices for TTS
- Speaker/room names
- Device nicknames
- Anything environment-specific

## Examples

```markdown
### Cameras

- living-room → Main area, 180° wide angle
- front-door → Entrance, motion-triggered

### SSH

- home-server → 192.168.1.100, user: admin

### TTS

- Preferred voice: "Nova" (warm, slightly British)
- Default speaker: Kitchen HomePod
```

## Why Separate?

Skills are shared. Your setup is yours. Keeping them apart means you can update skills without losing your notes, and share skills without leaking your infrastructure.

---

Add whatever helps you do your job. This is your cheat sheet.

## Pangolinfo CLI Skills（2026-06-12）

6 个已安装 skill，路径 `/workspace/skills/pangolinfo-*/scripts/`

**运行前置**：
```bash
export PANGOLINFO_API_KEY='<JWT>'
export HTTP_PROXY=http://192.168.100.1:7890
export HTTPS_PROXY=http://192.168.100.1:7890
```

**主要脚本**：
- scraper: `pangolinfo.py --q "keyword" --site amz_us` 或 `--content ASIN`
- niche: `pangolinfo.py --api category-search --keyword "xxx"`
- serp: `pangolinfo.py --q "xxx" --mode serp`
- product-explorer / listing-optimization / competitor-radar: 各自 scripts/ 下 amazon_scraper.py / amazon_niche.py 等

**注意**：scrapeapi.pangolinfo.com 不走代理超时，必须挂 HTTP_PROXY。

## Playwright

### 安装浏览器（Docker 环境）

必须带 `PLAYWRIGHT_BROWSERS_PATH` 环境变量，否则路径对不上：

```bash
docker exec openclaw-foreign-trade sh -lc \
  'PLAYWRIGHT_BROWSERS_PATH=/home/node/.openclaw/ms-playwright npx -y playwright install chromium'
```

## Related

- [Agent workspace](/concepts/agent-workspace)

### Proxy

- **地址**: `http://192.168.100.1:7890`
- **环境变量**:
  ```
  HTTP_PROXY=http://192.168.100.1:7890
  HTTPS_PROXY=http://192.168.100.1:7890
  ALL_PROXY=http://192.168.100.1:7890
  http_proxy=http://192.168.100.1:7890
  https_proxy=http://192.168.100.1:7890
  all_proxy=http://192.168.100.1:7890
  NO_PROXY=127.0.0.1,localhost,::1,cliproxyapi,ragflow,openclaw-gateway,openclaw-foreign-trade,192.168.100.1,192.168.100.135
  no_proxy=127.0.0.1,localhost,::1,cliproxyapi,ragflow,openclaw-gateway,openclaw-foreign-trade,192.168.100.1,192.168.100.135
  ```
- **curl**: `curl -x http://192.168.100.1:7890 <URL>`
- **Playwright**: `--proxy-server http://192.168.100.1:7890`
- **内网服务不走代理**: RAGFlow, cliproxyapi, localhost, OpenClaw
- **直连失败时自动重试代理**


<!-- tools-governance:start -->
## MCPHub / Tools 使用规则（2026-06-06）

- 默认只使用当前 agent 配置的单一 MCPHub 入口：Amazon → `AmazonOpenClaw`，OpenClaw Gateway → `OpenClawGateway`，Hermes → `Hermes`。
- 不要绕过 MCPHub 私自新增重复 MCP server；需要新工具时先加到 MCPHub 对应分组，再由 agent 现有 MCP server 继承。
- filesystem 工具只能操作本 agent 授权路径；不要用其他 agent 的 filesystem 改私有配置，除非用户明确要求跨实例维护。
- terminal-control、desktop/SSH、Docker、systemd、compose、文件删除、批量改写、镜像更新、网络代理修改属于高风险工具；执行前必须先说明目标、影响范围和回滚/备份点。
- Web/search/fetch/playwright/markitdown/scrapling/context7 在中国网络环境下优先使用已配置代理或 MCPHub 内的固定运行时，不要让每个 agent 自己重复下载依赖。
- 工具返回结果和向量记忆冲突时，以工具实时结果为准；工具失败不能伪装成功，要报告错误和下一步验证命令。
<!-- tools-governance:end -->