AugmentClaude

BOSS Agent CLI

Local assistant CLI for BOSS job platform with read-only, user-triggered operations.

Installation

  1. Make sure Claude is on your device and in your terminal.

    Skills load from ~/.claude/skills/ when Claude Code starts up — so you need it on your machine first. If you don't have it yet, install it once with the command below, then run claude in any terminal to verify.

    One-time setup
    npm i -g @anthropic-ai/claude-code

    Already have it? Skip ahead.

  2. Paste into Claude Code or into your terminal.

    This copies the whole skill folder into ~/.claude/skills/skill-can4hou6joeng4/ — the SKILL.md plus any scripts, reference docs, or templates the skill ships with. Safe default: works for every skill.

    Faster alternative (instruction-only skills)

    Skips the clone and grabs only the SKILL.md file. Don't use this if the skill ships Python scripts, reference markdowns, or asset templates — they won't be downloaded and the skill will fail when it tries to load them.

    Quick install (SKILL.md only)
    Sign up to copy
  3. Restart Claude Code.

    Quit and reopen Claude Code (or any other agent that loads from ~/.claude/skills/). New skills are picked up on startup.

  4. Just ask Claude.

    Skills auto-activate when your request matches the skill's description — no slash command needed. Trigger phrases live in the skill's own frontmatter; you can read them in the “What this skill does” section above.

Prefer to read the source first? Open on GitHub.

When Claude uses it

Claude activates this skill based on the context of your request.

What this skill does

boss-agent-cli

AI Agent 专用的 BOSS 直聘本地辅助 CLI。默认低风险模式只做本地辅助、只读优先、用户主动触发,不自动触达、不批量操作、不抓取平台数据。

Install

Skills CLI

npx skills add can4hou6joeng4/boss-agent-cli

CLI

uv tool install boss-agent-cli
patchright install chromium  # only needed for browser-assisted login

First Minute

按这个顺序跑,不需要先读完整 README:

boss doctor
boss schema
boss status
boss login   # only if status reports AUTH_REQUIRED / AUTH_EXPIRED

完成标准:

  • boss doctor 返回 ok=true,或给出可执行的 error.recovery_action
  • boss schema 返回当前命令、参数、平台和错误码自描述。
  • boss status 返回当前登录态;未登录时由用户主动执行 boss login

Safe Candidate Loop

求职者侧默认只推荐这个低风险闭环:

boss schema
boss search "Golang" --city 广州 --welfare "双休,五险一金"
boss detail <security_id>
boss shortlist add <security_id> <job_id>

规则:

  • 先调用 boss schema,不要硬编码完整命令表。
  • 只解析 stdout 的 JSON 信封;stderr 只当日志和进度信息。
  • ok=false 时读取 error.codeerror.recovery_action,不要猜测恢复步骤。
  • --welfare 是核心筛选能力,适合用于低风险职位搜索和本地整理。
  • 投递、打招呼、沟通、交换联系方式等动作回到平台官网由用户手动完成。

Recruiter Boundary

招聘者侧默认只保留低风险职位管理入口:

boss schema
boss hr jobs list
boss hr jobs online <job_id>
boss hr jobs offline <job_id>

候选人搜索、投递申请、在线简历、附件简历、聊天记录、消息回复、联系方式交换等链路涉及个人信息或平台关系写入,默认低风险模式会返回 COMPLIANCE_BLOCKED。遇到这类结果时,停止自动化,回到官方页面由用户手动处理。

Output Contract

所有命令都遵守同一 stdout JSON 信封:

{
  "ok": true,
  "schema_version": "1.0",
  "command": "search",
  "data": [],
  "pagination": null,
  "error": null,
  "hints": {}
}
  • stdout: 只输出 JSON 信封。
  • stderr: 只输出日志和进度,受 --log-level 控制。
  • exit code 0: ok=true
  • exit code 1: ok=false,必须读取 error.codeerror.recoverableerror.recovery_action

Recovery

优先按错误信封恢复:

Error codeAgent action
AUTH_REQUIRED / AUTH_EXPIRED / TOKEN_REFRESH_FAILED让用户主动执行 boss login,然后再跑 boss status
RATE_LIMITED / NETWORK_ERROR等待后按 recovery_action 重试
INVALID_PARAM修正参数后重试,例如城市、页码、福利关键词
AI_NOT_CONFIGURED / AI_API_ERROR / AI_PARSE_ERRORrecovery_action 配置或重试 AI 能力
ACCOUNT_RISK / COMPLIANCE_BLOCKED停止自动化,使用平台官网或官方页面手动完成

不要把登录降级、浏览器通道或 CDP 当作规避风控的手段。

Docs

License

MIT

Related skills