boss-crawler-skill

A Claude Code skill that collects BOSS Zhipin vacancies, matches them against a resume and can apply to the best ones.

  • 121 ★ on the repository
  • Last commit
Skill

Broad access

Why this level

  • The skill logs in and sends applications to employers on your behalf.
  • Scraping the platform and imitating human actions may violate platform terms and lead to account blocking.
All reasons and checks

zhansan379/boss-crawler-skill

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/boss-crawler-skill

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Broad accessThe tool can change data, run commands or handle secrets.

  • The skill logs in and sends applications to employers on your behalf.
  • Scraping the platform and imitating human actions may violate platform terms and lead to account blocking.

This is third-party code. Review the repository files before installing.

What it does

The skill walks a job seeker through a chain on the BOSS Zhipin platform. It first parses the resume and infers search parameters, then collects vacancies through DrissionPage and scores them with rules and a language model. It then builds an HTML report, prepares a resume image without personal data tailored to a vacancy, writes individual greetings and, if wanted, sends applications. There are two modes. In Claude Code the `/boss-crawler` command guides through stages with confirmations, while on the command line the same scripts run with all parameters at once. The model is connected through your own endpoint: Anthropic Messages by default, with OpenAI-compatible formats, DeepSeek, Qwen, Zhipu and local ollama and vLLM also supported.

Who it is for. Job seekers on the Chinese labour market with a BOSS Zhipin account who want to filter vacancies by resume and apply selectively.

Good fit when

  • There are hundreds of vacancies and only those fitting your stack and experience should remain.
  • Each application needs an individual greeting based on the resume and the vacancy text.
  • The resume must be adapted to a vacancy and an image without contacts should be sent.
  • The search is repeated with saved parameters without manual setup.

Not a fit when

  • You want bulk platform data collection for resale: the author explicitly forbids that use.
  • The account matters to you: the platform may restrict automated actions.
  • You need to work without a Chinese account or on another platform.

Example request

/boss-crawler: take my resume, find Python vacancies in Shenzhen, show the report and apply only to high-match vacancies after my confirmation.

Limitations

Automated login and applications run from your BOSS Zhipin account, so the platform may block it for rules or signs of automation. The author states the project is for study and asks not to violate the platform terms. The skill targets Chinese-speaking users and replies in Chinese. No licence is declared in the repository. Python 3.8 or newer, Chrome and a model key in `assets/llm_config.json` are required. Resumes and vacancies are sent to the chosen model, and the skill waits 3 to 5 seconds between applications. Last commit September 2026.

How to disable. Delete the `~/.claude/skills/boss-crawler` directory, then check the saved login state and the keys in `assets/llm_config.json`. The session can be revoked in the BOSS Zhipin account settings.

FAQ

How do I install the skill?

Ask Claude Code to install the repository from the link or clone it into `~/.claude/skills/boss-crawler`, install dependencies and fill `assets/llm_config.json`.

Which models work?

The Anthropic Messages protocol by default. OpenAI-compatible endpoints also work: DeepSeek, Qwen, Moonshot, Zhipu, Volcengine, local ollama and vLLM.

What does `llm_check.py` do?

It checks model settings and shows the key masked. The `--no-call` flag checks configuration only, with no requests and no cost.

Can it run without Claude Code?

Yes. The same scripts run from the command line with one set of parameters, and the search can be repeated on a schedule.

Does the skill require Chinese for questions?

The skill asks and answers in Chinese as set in `SKILL.md`. Commands, paths and script names stay in English.

Discussion

No questions or reviews yet.

Skills

Skills for HR and recruiters: role requirements, sourcing on BOSS Zhipin and Liepin, resume review, interviews and a candidate ledger

SkillBroad access109Repository stars

An MCP server and CLI for the BOSS job platform: search vacancies, filter by conditions and shortlist with JSON output

MCP serverExtended access2.2KRepository stars

An agent on your Claude or Codex subscription finds jobs, tailors resumes and applies, with every status shown on a web dashboard.

PluginBroad access109Repository stars

Skill and CLI for job applications: fills Greenhouse, Ashby and Lever forms and logs only confirmed submissions

SkillBroad access173Repository stars

A curated list of Claude skills and plugins, plus Composio's own automation skills for 78 SaaS apps

SkillBroad access78.9KRepository stars

Eight open source AI employees for business roles: 60 scheduled routines, work in your own browser and a morning brief, on Claude Code and other agents

PluginBroad access851Repository stars

README in short

One driver `scripts/pipeline.py` runs nine stages from resume parsing to the report, with applying as a separate step. All model inference happens inside the commands.

README badge

Report a problem

Foxx AIboss-crawler-skill
Foxx AI

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.