[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-92528":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":8,"htmlUrl":8,"language":9,"languages":8,"totalLinesOfCode":8,"stars":10,"forks":11,"watchers":12,"openIssues":13,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":14,"stars7d":14,"stars30d":15,"stars90d":14,"forks30d":14,"starsTrendScore":14,"compositeScore":16,"rankGlobal":8,"rankLanguage":8,"license":17,"archived":18,"fork":18,"defaultBranch":19,"hasWiki":18,"hasPages":18,"topics":20,"createdAt":8,"pushedAt":8,"updatedAt":21,"readmeContent":22,"aiSummary":23,"trendingCount":14,"starSnapshotCount":14,"syncStatus":24,"lastSyncTime":25,"discoverSource":26},92528,"cn-company-radar-toolbox","zexianzhang36-prog\u002Fcn-company-radar-toolbox","zexianzhang36-prog",null,"Python",58,5,51,1,0,6,39.93,"GNU Affero General Public License v3.0",false,"main",[],"2026-07-22 04:02:06","# CN Company Radar Toolbox · 中国求职公司雷达工具箱\n\n> Find better companies before sending more resumes.\n> 先发现真正值得投递的公司，再去匹配岗位、优化简历、推进沟通。\n\n[![License: AGPL v3](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-AGPL%20v3-blue.svg)](LICENSE)\n[![Python 3.12](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fpython-3.12-blue.svg)](.python-version)\n[![FastAPI](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FFastAPI-REST%20%2B%20MCP-009688.svg)](backend\u002Fpyproject.toml)\n[![React](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FReact-18%20%2B%20TypeScript-61DAFB.svg)](frontend\u002Fpackage.json)\n[![Docker](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fdocker-compose-2496ED.svg)](docker-compose.yml)\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fassets\u002Fscreenshots\u002F01-landing.png\" alt=\"CN Company Radar landing page\" width=\"92%\">\n\u003C\u002Fp>\n\n## Read This First \u002F 选择适合你的版本\n\n| Reader | Start here |\n| --- | --- |\n| 求职者、应届生、社招、转行、来中国工作的外国朋友 | [候选人阅读版 README.candidate.md](README.candidate.md) |\n| 开发者、面试官、招聘方、开源贡献者 | [开发者阅读版 README.developer.md](README.developer.md) |\n\nCN Company Radar Toolbox 是一个面向中国求职场景的本地化 AI 工具箱。它把企查查、上市公告、官方认定名单、政策、招投标、招聘岗位、GitHub 技术活跃、网页抓取、个人收藏库等信号，统一成 AI 可以调用、用户可以追溯的结构化证据。\n\nThis project is a local-first AI job-search cockpit for China. It helps job seekers discover good companies, understand why they are good, match roles with resumes, and keep the whole process evidence-based instead of rumor-based.\n\n它不是一个“海投机器”。它想解决的是信息不对称：很多好机会不在热搜里，不在大厂榜单里，也不一定被刚毕业的学生、转行的人、社招朋友，或者第一次来中国工作的外国朋友看见。这个工具希望把分散的公开信号聚合起来，让每个人都能更有尊严、更有方向地找工作。\n\n## Why Star This Project \u002F 为什么值得 Star\n\nIf this repo helps you believe job search can be more transparent, evidence-based, and fair, a Star helps more people find it.\n\n- For job seekers: it turns scattered company\u002Fjob information into a practical decision dashboard.\n- For AI builders: it is a complete local-first AI product with REST, MCP, React, CLI, scoring, evidence, and browser userscript automation.\n- For students and career changers: it is a real-world full-stack portfolio project, not just another demo.\n- For the open-source community: it keeps the core principle clear: facts first, evidence traceable, human in control.\n\n## Product Preview \u002F 产品预览\n\n| Discover better companies | Verify one company deeply |\n| --- | --- |\n| ![Multi-Boards](docs\u002Fassets\u002Fscreenshots\u002F02-multi-boards.png) | ![Company Detail](docs\u002Fassets\u002Fscreenshots\u002F03-company-detail.png) |\n| Multi-Boards rank companies by small gems, big players, industry, region, and real jobs. | Company Detail shows business profile, AI score, evidence, recognition tags, and jobs. |\n\n| Follow up BOSS jobs | Match resumes to roles |\n| --- | --- |\n| ![BOSS Pipeline](docs\u002Fassets\u002Fscreenshots\u002F04-boss-pipeline.png) | ![Resume Matcher](docs\u002Fassets\u002Fscreenshots\u002F05-resume-matcher.png) |\n| BOSS pipeline keeps greetings, replies, AI assistance, and manual resume delivery separate. | Resume matcher connects JD, ATS keywords, AI tailoring, and BOSS greetings. |\n\n| Inspect every data capability |\n| --- |\n| ![Capabilities](docs\u002Fassets\u002Fscreenshots\u002F06-capabilities.png) |\n| Every tool source declares cost, API-key requirement, latency, freshness, evidence strength, and best-use cases. |\n\n## Website Tour \u002F 网页导览\n\n| Page | What it does |\n| --- | --- |\n| 首页 \u002F Landing | Explains the pain point: job seekers do not just need more jobs; they need better company discovery. |\n| 多榜单 \u002F Multi-Boards | Small gems, big players, industries, regions, jobs, and natural-language search in one board. |\n| 公司查询 \u002F Company Lookup | Resolve Chinese company names into stable IDs and evidence timelines. |\n| 简历 \u002F Resume | Upload resumes, parse to Markdown, check JD fit, tailor content, generate greetings. |\n| 求职投递 \u002F Job Apply | Track BOSS jobs from greeting to reply to manual resume delivery. |\n| 我的库 \u002F My Library | Save searched\u002Fapplied companies and jobs into a reusable local candidate pool. |\n| 官方信号 \u002F Official Signals | One-click access to recognized SMEs, policies, bidding records, and AI filings. |\n| 能力目录 \u002F Capabilities | Inspect every data adapter, status, cost, API-key need, latency, and evidence strength. |\n| 调用工具 \u002F Run Tool | Run any registered capability with JSON and inspect structured evidence. |\n| Tool Probe | Run source health checks and compare declared status with live behavior. |\n| AI 助手 \u002F Assistant | Ask natural-language career questions and let AI orchestrate the toolbox. |\n\n## Mission \u002F 项目愿景\n\n让每一个认真找工作的人，少一点盲投和焦虑，多一点证据和选择。\n\nLet every serious job seeker spend less time guessing, and more time choosing with clarity.\n\nCN Company Radar 关注两件事：\n\n- 帮求职者看见被忽视的好公司：小而美、专精特新、真实在招、技术活跃、有政策\u002F招投标\u002F产品落地信号。\n- 帮踏实做事的公司被合适的人看见：不是所有好工作都来自大厂，也不是所有好公司都会做营销。\n\n## Who Is It For \u002F 适合谁\n\n| 人群 | 你会用它解决什么问题 |\n| --- | --- |\n| 应届生 \u002F 秋招同学 | 找到校招友好、真实在招、不是只有大厂才值得投的公司。 |\n| 社招朋友 | 按城市、行业、薪资、公司阶段和证据强度筛选更靠谱的机会。 |\n| 转行者 | 先理解赛道、公司和岗位要求，再把简历改到目标岗位语言里。 |\n| 来中国工作的外国朋友 \u002F International friends in China | Understand China-specific company signals: official recognition, filings, procurement, hiring signals, and local company profiles. |\n| 独立开发者 \u002F AI learners | 学习一个完整的 FastAPI + React + AI orchestration + browser userscript 工程。 |\n\n## Core Idea \u002F 核心理念\n\n```\n多源信号 -> 标准化证据 -> 公司评分与分榜 -> 岗位匹配 -> 简历优化 -> 求职跟进\n          Evidence first        AI as brain       Human in control\n```\n\n工具箱只尽量产出客观事实、结构化信号和可追溯证据。\n“哪家公司值得投”“哪份简历更适合”“要不要继续沟通”这些判断，由用户、配置和 AI 大脑共同完成。\n\n## Product Map \u002F 功能地图\n\nOpen `http:\u002F\u002F127.0.0.1:8770\u002F` after startup and you get a complete local job-search workspace:\n\n### 1. Multi-Boards \u002F 多榜单公司雷达\n\n一次发现评分，分多个视角看公司和岗位：\n\n- 小而美榜：未上市中小主体、专精特新、小巨人、潜在独角兽、应届友好信号。\n- 大而强榜：上市公司、独角兽、大厂、行业头部。\n- 行业榜：按赛道聚合公司，例如大模型、AI 安全、信息安全、企业服务等。\n- 地区榜：按城市聚合机会，适合看本地就业与迁移选择。\n- 岗位榜：真实在招岗位，区分校招\u002F社招，展示发布时间和时效。\n- 自然语言筛选：例如 `大模型 杭州`，自动识别方向和城市，返回匹配公司与岗位。\n\n### 2. Company Detail \u002F 公司详情\n\n从榜单下钻到公司详情，看到：\n\n- 公司主营业务和简介。\n- AI 评分、推荐动作、命中关键词和评分依据。\n- 企业性质、规模、融资阶段、官方认定标签。\n- 招聘岗位、校招\u002F社招、发布时间、岗位来源。\n- `AI 重新整理` 和 `企查查权威核验`，缓存优先，避免无意义消耗 API 次数。\n\n### 3. Company Lookup \u002F 公司查询与实体消歧\n\n输入公司名、统一社会信用代码或城市，自动解析成稳定的 `company_id`，再拉取实体属性和证据时间线。\n这解决了中国公司常见的简称、品牌名、主体名混用问题。\n\n### 4. Official Signals \u002F 官方信号\n\n一键调用四类更可信的一手信号：\n\n- 政府认定名单：专精特新“小巨人”、独角兽、单项冠军等。\n- 政策信息：国务院政策库，理解赛道顺风和监管方向。\n- 招投标：政府采购和中标记录，验证真实交付能力。\n- 生成式 AI 备案：判断 AI 产品是否真实上线并接受监管。\n\n### 5. Resume Library & Matcher \u002F 简历库与岗位匹配\n\n把“找到岗位”接到“提高命中率”：\n\n- 上传 PDF、图片、Markdown、TXT 简历。\n- 支持文本提取和多模态 OCR 解析。\n- 管理主简历、AI 优化简历、行业版本。\n- 粘贴 JD 或选择已采集 BOSS 岗位。\n- 本地 ATS 关键词覆盖率，无 API Key 也能用。\n- 配置 AI Key 后可做深度匹配、差距分析、定制改写和 BOSS 招呼语生成。\n\n### 6. BOSS Job Pipeline \u002F BOSS 求职投递跟进\n\n这是工具箱的“手脚”，用于把真实岗位回流到本地求职流水线：\n\n- BOSS userscript 在用户自己的浏览器登录态下运行。\n- 岗位采集回流到本地 `data\u002Fimports\u002Fboss_jobs.jsonl` 和岗位榜。\n- 按月薪、年薪、城市、关键词筛选岗位。\n- 看板阶段：待开始 -> 已打招呼·监测中 -> HR 已回复 \u002F AI 回复中 -> 待投递简历。\n- 批量打招呼有护栏：每日上限、随机延时、去重、脚本状态诊断。\n- 打招呼不等于投递简历。最终是否发简历，由用户人工决定。\n\n### 7. My Library \u002F 我的库\n\n每次在发现页搜索到的公司和岗位会自动沉淀到本地库：\n\n- 公司库：记录你看过、搜索过、投递过的公司。\n- 岗位库：记录候选岗位、来源和后续状态。\n- 库里的公司会重新进入榜单评分，让个人候选池越用越准。\n\n### 8. Capabilities \u002F 能力目录\n\n当前注册 20 个数据源能力，其中本地实测页面显示 14 个 `LIVE`、6 个 `STUB`。每个能力都标注：\n\n- 输入参数。\n- 是否需要 API Key。\n- 成本、延迟、时效、证据强度。\n- 擅长什么，不适合什么。\n\n已接入或登记的能力包括：`ai_filing`、`akshare`、`bidding`、`cninfo`、`crawl4ai`、`github`、`jobs`、`manual_import`、`policy`、`qcc`、`qualified_smes`、`saved_companies`、`vc_portfolio`、`xiaohongshu`，以及若干待做实 stub。\n\n### 9. Run Tool & Tool Probe \u002F 调用工具与探针\n\n开发者和高级用户可以直接验证每个数据源：\n\n- Run Tool：选择能力源，输入 JSON query，查看结构化数据和 evidence。\n- Tool Probe：实跑所有源，对照声明状态检查可用性、延迟、样例输出和是否值得继续建设。\n\n### 10. AI Assistant \u002F AI 助手\n\n用自然语言驱动工具箱：\n\n- “帮我找杭州做 AI Agent 的早期公司，重点看技术活跃。”\n- “有哪些做数据安全、且中过政府采购标的公司？”\n- “对比专精特新小巨人里适合应届生投递的 AI 公司。”\n\n支持 `api`、本地 `claude_cli`、本地 `codex_cli` 等后端。会话记忆保存在本地，不是云端黑箱。\n\n## Typical Workflow \u002F 一条真实求职流水线\n\n```text\n1. 用多榜单发现公司\n   Small Gems \u002F Big Players \u002F Industry \u002F Region \u002F Jobs\n\n2. 下钻公司详情\n   看主营业务、证据、岗位、评分依据、是否适合应届或当前职业阶段\n\n3. 定向采集岗位\n   用 BOSS userscript 把真实岗位回流到本地\n\n4. 匹配并优化简历\n   ATS keyword coverage -> AI deep match -> tailored resume -> greeting\n\n5. 谨慎推进沟通\n   打招呼 -> 监测回复 -> AI 辅助回复 -> 用户决定是否投简历\n```\n\n## Tech Stack \u002F 技术栈\n\n- Backend: Python 3.12, FastAPI, Pydantic v2, SQLite, Typer CLI, fastmcp.\n- Frontend: Vite, React 18, TypeScript, Tailwind CSS, shadcn-style components, lucide icons.\n- AI layer: provider-agnostic brain backend, local CLI or API mode.\n- Browser side: ScriptCat\u002FTampermonkey userscript for BOSS job capture and local-only workflow automation.\n- Data: local `data\u002Fseed` sample data, local `data\u002Fimports` runtime data, configurable scoring in `configs\u002F*.yaml`.\n\n```text\nadapters \u002F data sources\n        -> ToolResult \u002F Evidence\n        -> FastAPI REST + MCP server\n        -> React cockpit \u002F AI assistant \u002F CLI\n        -> local job-search workflow\n```\n\n## Quickstart \u002F 快速开始\n\n### Option A: Docker\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fzexianzhang36-prog\u002Fcn-company-radar-toolbox\ncd cn-company-radar-toolbox\ncp .env.example .env\ndocker compose up -d --build\n```\n\nOpen:\n\n```text\nhttp:\u002F\u002F127.0.0.1:8770\u002F\n```\n\n### Option B: Local bootstrap\n\nLinux \u002F macOS \u002F Git Bash:\n\n```bash\ncp .env.example .env\n.\u002Fbootstrap.sh --serve\n```\n\nWindows PowerShell:\n\n```powershell\nCopy-Item .env.example .env\n.\\bootstrap.ps1 -Serve\n```\n\nThe repository ships with `data\u002Fseed\u002F` sample data, so the boards have content on first launch.\n\n## BOSS Userscript \u002F BOSS 油猴脚本\n\nInstall ScriptCat or Tampermonkey, then open:\n\n```text\nhttp:\u002F\u002F127.0.0.1:8770\u002Fboss\u002Fboss-toolbox.user.js\n```\n\nSee [docs\u002FBROWSER_TOOLS.md](docs\u002FBROWSER_TOOLS.md) for installation, panel buttons, diagnostics, and troubleshooting.\n\n## Configuration \u002F 配置\n\nMost basic features work without API keys. Optional keys unlock stronger AI and verification:\n\n- `DEEPSEEK_API_KEY` or other LLM keys: AI assistant, deep resume match, resume tailoring.\n- `DASHSCOPE_API_KEY`: multimodal resume OCR.\n- `QCC_TOKEN`: Qichacha authoritative verification.\n- `QCC_MAX_CALLS`: local quota guardrail.\n\nConfig is stored in local `.env` only. Do not expose this service directly to the public internet without authentication.\n\n## Responsible Use \u002F 负责任使用声明\n\nThis project is built for personal job seeking, learning, research, and technical exchange only.\n\n- 本项目不构成投资建议、法律建议、职业承诺或雇佣承诺。榜单和评分只是辅助判断，请自行核验。\n- 抓取、采集、浏览器脚本相关能力只应用于你自己账号、你自己可见的数据和个人求职辅助。\n- 请遵守目标网站服务条款、robots、平台规则和当地法律法规。\n- 不得用于商业化爬取、批量骚扰 HR、账号滥用、绕过权限、盗取数据、出售数据或任何未授权行为。\n- 真实抓取数据可能包含 HR、岗位、聊天、简历等个人信息，默认已被 `.gitignore` 排除。不要公开提交或发布。\n- 本服务默认是单用户本地工具，API 没有鉴权。公网部署必须自行加认证、HTTPS、访问控制和配额保护。\n\n### BOSS \u002F XSS \u002F Userscript Notice\n\n仓库中的 BOSS 相关浏览器脚本包含页面注入、网络钩子、自诊断和本地自动化实验代码。这里的 “XSS \u002F injection” 语境只用于安全学习、浏览器 userscript 研究和个人求职流程自动化交流。\n\nIt is not provided as an exploit kit. It must not be used to attack websites, steal credentials, access data beyond your own account, bypass authorization, or create spam\u002Fabuse traffic. The user remains fully responsible for how they install, configure, and run it.\n\n## Documentation \u002F 文档\n\n| 文档 | 内容 |\n| --- | --- |\n| [docs\u002FDEPLOYMENT.md](docs\u002FDEPLOYMENT.md) | Docker、本地脚本、手动部署、迁移说明 |\n| [docs\u002FCONFIGURATION.md](docs\u002FCONFIGURATION.md) | `.env` 和 `configs\u002F*.yaml` 配置说明 |\n| [docs\u002FBROWSER_TOOLS.md](docs\u002FBROWSER_TOOLS.md) | BOSS userscript 安装、面板、采集路径、故障排查 |\n| [DESIGN.md](DESIGN.md) | 产品设计、架构、数据模型、路线图和合规思路 |\n| [SECURITY.md](SECURITY.md) | 密钥、数据、部署和漏洞报告策略 |\n| [CONTRIBUTING.md](CONTRIBUTING.md) | 贡献指南 |\n\n## License \u002F 许可\n\nThis project is licensed under [AGPL-3.0](LICENSE).\n\n如果你修改后作为网络服务对外提供，必须按 AGPL 要求开源相应修改。若需要闭源 SaaS、企业内部分发或其他商业使用方式，请联系作者获取商业授权。\n\nCommercial license \u002F dual-license inquiries: `jackluo217330@gmail.com`\n\n## Contributing \u002F 参与贡献\n\n欢迎 PR、Issue 和真实使用反馈。建议优先贡献：\n\n- 更可靠的数据源 adapter。\n- 更好的中文\u002F英文求职场景提示词。\n- 更清晰的合规边界和安全护栏。\n- 面向应届、社招、转行、外国朋友的本地化求职指南。\n\n核心原则：facts first, evidence traceable, human in control.\n","CN Company Radar Toolbox 是一个面向中国求职者的本地化AI辅助决策工具箱，帮助用户基于多源公开数据（如企业信用信息、招聘平台岗位、政策认定名单、技术活跃度等）系统性发现和评估值得投递的公司。核心功能包括多维度公司雷达评分、证据可追溯的AI分析、BOSS直聘流程集成、简历-岗位智能匹配及本地优先的全栈架构（FastAPI后端 + React前端 + Docker部署）。适用于应届生、社招人士、转行者及在华外籍求职者，解决信息不对称问题，支持理性、透明、有依据的职业决策。",2,"2026-07-09 02:30:14","CREATED_QUERY"]