← 返回
未分类 Key 中文

Mibatt Dietu Cli Agent

Use this skill when you need to operate Dietu through its official CLI for A-share market queries, strategy screening, decision workflows, or agent automatio...
在需要通过 Dietu 官方 CLI 进行 A 股市场查询、策略筛选、决策工作流或代理自动化等操作时,使用此技能。
kenee kenee 来源
未分类 clawhub v1.0.1 2 版本 100000 Key: 需要
★ 0
Stars
📥 388
下载
💾 1
安装
2
版本
#latest

概述

Dietu CLI Agent

Use this skill when the task should be completed through the official dietu CLI instead of web UI clicks or backend-only scripts.

Typical fit:

  • Query A-share market overviews, stock snapshots, indices, or search results
  • Run strategy lists and screening commands
  • Pull decision, trading, and review data from Dietu
  • Set up CLI authentication for human users or automation
  • Produce machine-readable CLI output for another agent or script

Base URL

Default API endpoint: https://dietu-api.mibatt.com

Override via --base-url, DIETU_BASE_URL, or profile config.

Quick rules

  • Prefer the public CLI package: npm i -g @mibatt/dietu@latest
  • For human login, use dietu auth login
  • For automation, prefer a PAT in DIETU_ACCESS_TOKEN
  • For agent consumption, prefer --format json
  • For user-facing pasted output, prefer --format markdown
  • For quick terminal inspection, prefer --format table
  • Use real ts_code symbols such as 600519.SH or 000001.SZ
  • If you do not know the command surface, run dietu schema
  • If connectivity or auth looks suspicious, run dietu doctor

Authentication

Read references/auth.md when the task involves login, token sourcing, base URL switching, or PAT usage.

Important:

  • The primary interactive path is GitHub device flow
  • The recommended automation path is PAT from Settings / Access Tokens
  • Do not invent outdated commands such as dietu ping or dietu mcp start

Command recipes

Read references/commands.md when you need concrete command sequences for:

  • market
  • research
  • decision
  • trading
  • review
  • schema and doctor

Working style

  • Start with the smallest command that can answer the task
  • Add --format json before doing downstream parsing or summarization
  • Preserve the user's chosen --base-url, --profile, or token source
  • For commands that require an account, explicitly provide --account
  • If the task spans multiple CLI calls, keep outputs structured and consistent

Non-goals

  • Do not call internal Python scripts as the primary interface when the CLI already exposes the capability
  • Do not assume browser-only login or password login is the preferred path
  • Do not treat short-lived JWT copied from local profile as the recommended long-term automation credential

版本历史

共 2 个版本

  • v1.0.1 当前
    2026-05-08 12:59 安全 安全
  • v0.1.0
    2026-05-07 10:28 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

ai-agent

self-improving agent

pskoett
记录自身发现以实现自我改进的技能
★ 4,163 📥 933,447
ai-agent

Self-Improving + Proactive Agent

ivangdavila
自我反思+自我批评+自我学习+自组织记忆。智能体评估自身工作、发现错误并持续改进。
★ 1,441 📥 328,185
ai-agent

Agent Browser

rez0
用于 AI 代理的浏览器自动化 CLI。当用户需要与网站交互(包括浏览页面、填写表单、点击按钮、截图等)时使用。
★ 865 📥 343,468