{"_id":"@aiwaretop/gaokao-volunteer-research","name":"@aiwaretop/gaokao-volunteer-research","dist-tags":{"latest":"0.3.1"},"versions":{"0.3.1":{"name":"@aiwaretop/gaokao-volunteer-research","version":"0.3.1","description":"Official-source-first Codex/Avatanel skill for Chinese Gaokao volunteer research.","license":"MIT","author":{"name":"aiware"},"homepage":"https://github.com/HackSing/gaokao-volunteer-research#readme","repository":{"type":"git","url":"git+https://github.com/HackSing/gaokao-volunteer-research.git"},"bugs":{"url":"https://github.com/HackSing/gaokao-volunteer-research/issues"},"publishConfig":{"access":"public","registry":"https://registry.npmjs.org/"},"keywords":["codex-skill","avatanel-skill","gaokao","college-admissions","education","research"],"bin":{"gaokao-toolkit":"scripts/gaokao_toolkit.py"},"scripts":{"test":"python3 scripts/gaokao_toolkit.py regression --cases tests/regression-cases.json && python3 scripts/gaokao_toolkit.py validate-package --dir tests/fixtures/sample-package --province 广东 --year 2026 --category 物理类 && python3 scripts/gaokao_toolkit.py parse-table --input tests/fixtures/guangdong-score-table.md --out /tmp/gaokao-score-table.json --kind score --province 广东 --year 2026 --category 物理类 && python3 -m py_compile scripts/gaokao_toolkit.py","pack:check":"npm pack --dry-run"},"gitHead":"b586ccdbdbf4cdb8b9822c1a2f825f08f7fe1779","_id":"@aiwaretop/gaokao-volunteer-research@0.3.1","_nodeVersion":"25.8.0","_npmVersion":"11.11.0","dist":{"integrity":"sha512-2Ry2UhmB3OoZ5ex3BLLZmeTj+CKi9qMQBeuVkmsvVDRQQ8EOLmdDYqa9zfhaoTdQs9bVJAkapreMoV4dwLf8kQ==","shasum":"3c39fddcb2755c021b8ec6fa99059521d9455889","tarball":"https://registry.npmjs.org/@aiwaretop/gaokao-volunteer-research/-/gaokao-volunteer-research-0.3.1.tgz","fileCount":22,"unpackedSize":63566,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEUCIQCnMlIckO+8tQZ16JN8Ls+aCnfV05sZyWS5YcOviauMOAIgdAus8j/3LbzA0fYDY/GCD9d7mn4q5p+hHXDN7OGcEEQ="}]},"_npmUser":{"name":"aiwarelife","email":"aiwarelife@gmail.com"},"directories":{},"maintainers":[{"name":"aiwarelife","email":"aiwarelife@gmail.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/gaokao-volunteer-research_0.3.1_1781441690036_0.34621118382951055"},"_hasShrinkwrap":false}},"time":{"created":"2026-06-14T12:54:49.870Z","0.3.1":"2026-06-14T12:54:50.171Z","modified":"2026-06-14T12:54:50.383Z"},"maintainers":[{"name":"aiwarelife","email":"aiwarelife@gmail.com"}],"description":"Official-source-first Codex/Avatanel skill for Chinese Gaokao volunteer research.","homepage":"https://github.com/HackSing/gaokao-volunteer-research#readme","keywords":["codex-skill","avatanel-skill","gaokao","college-admissions","education","research"],"repository":{"type":"git","url":"git+https://github.com/HackSing/gaokao-volunteer-research.git"},"author":{"name":"aiware"},"bugs":{"url":"https://github.com/HackSing/gaokao-volunteer-research/issues"},"license":"MIT","readme":"# 高考志愿研究 Skill\n\n`gaokao-volunteer-research` 是一个面向中国高考志愿研究的 Codex/Avatanel skill。它的目标不是“替你填志愿”，而是把志愿相关问题转成一套可查证、可复核、可交付的研究流程。\n\n它适合用于：\n\n- 高考志愿、大学志愿、选大学、选专业、冲稳保垫研究。\n- 一分一段、分数对应位次、本科线、专科线、批次线核对。\n- 招生章程、院校/专业信息、政策文档、高考真题来源整理。\n- 志愿参考前的口径校验和风险点整理。\n- 生成 `sources.md`、`data-check.md`、`candidate-matrix.md`、`family-brief.md`、`risk-notes.md` 等研究包文件。\n\n## 核心边界\n\n- 必须优先使用官方、当年、可追溯资料。\n- 缺少省份、年份、位次等关键口径时，不得直接给具体候选矩阵。\n- 不承诺录取结果，不预测录取概率，不声称掌握内部数据，不代替用户提交官方志愿系统。\n- 非官方榜单、短视频、截图、家长群表格只能作为线索，必须回到省考试院、阳光高考/阳光志愿、高校招生章程等来源核验。\n- 真题资料只整理来源、入口、年份、科目、卷别和授权状态，不大段转载非官方或未授权题面/答案解析。\n\n## 目录结构\n\n```text\ngaokao-volunteer-research/\n├── SKILL.md\n├── CHANGELOG.md\n├── README.md\n├── data/\n│   └── official-source-index.json\n├── examples/\n│   └── sample-run.md\n├── references/\n│   ├── source-policy.md\n│   ├── test-scenarios.md\n│   └── tooling.md\n├── scripts/\n│   └── gaokao_toolkit.py\n├── templates/\n│   ├── candidate-matrix.md\n│   ├── data-check.md\n│   └── family-brief.md\n└── tests/\n    ├── fixtures/\n    └── regression-cases.json\n```\n\n## 快速使用\n\n### 通过 npm 安装\n\n公开 npm 包发布后，可以通过 npm 安装：\n\n```bash\nnpm install @aiwaretop/gaokao-volunteer-research\n```\n\n如果要把确定性工具安装成全局命令：\n\n```bash\nnpm install -g @aiwaretop/gaokao-volunteer-research\n```\n\n全局安装后可直接使用：\n\n```bash\ngaokao-toolkit --help\ngaokao-toolkit index lookup --province 广东\n```\n\n如果你的 agent 支持从 npm 包目录加载 skill，把 npm 安装目录中的 `SKILL.md`、`references/`、`templates/`、`data/` 和 `scripts/` 作为 skill 目录使用。\n\n### 在 agent 中使用\n\n在支持 skill 的 agent 中直接说：\n\n```text\n使用 gaokao-volunteer-research。\n2026 年广东物理类 612 分，我不知道位次。先不要推荐学校，先帮我做 data-check：查一分一段、分数对应位次、本科线/专科线/批次线，并列出官方来源和下一步核验路径。\n```\n\n资料查证示例：\n\n```text\n使用 gaokao-volunteer-research。\n帮我整理 2026 年江苏高考真题、答案或官方试卷资料来源。不要转载题面，只列来源、URL、年份、科目/卷别和是否官方。\n```\n\n完整研究包示例：\n\n```text\n使用 gaokao-volunteer-research。\n2026 年广东，物理类，位次 18000，本科批，偏计算机/电子信息，公办优先，城市优先广州深圳珠三角。请生成 sources.md、data-check.md、candidate-matrix.md、family-brief.md 和 risk-notes.md。\n```\n\n## 工具层\n\n确定性工具入口是：\n\n```bash\npython3 scripts/gaokao_toolkit.py --help\n```\n\n常用命令：\n\n```bash\npython3 scripts/gaokao_toolkit.py index lookup --province 广东\npython3 scripts/gaokao_toolkit.py index verify --province 广东 --timeout 10\npython3 scripts/gaokao_toolkit.py parse-table --input tests/fixtures/guangdong-score-table.md --out /tmp/gaokao-score-table.json --kind score --province 广东 --year 2026 --category 物理类\npython3 scripts/gaokao_toolkit.py validate-package --dir tests/fixtures/sample-package --province 广东 --year 2026 --category 物理类\npython3 scripts/gaokao_toolkit.py regression --cases tests/regression-cases.json\n```\n\n工具能力：\n\n- `index`：查询和验证省级考试院、阳光高考等官方入口。\n- `snapshot`：保存政策、章程、PDF、真题来源页面等快照和 metadata。\n- `parse-table`：解析 CSV、TSV、Markdown、简单 HTML 表格。\n- `validate-package`：检查研究包文件、URL、口径和禁止表达。\n- `regression`：检查 skill 文档是否保留关键触发词和回归样例。\n\n## 推荐研究包\n\n默认研究包包含：\n\n```text\nsources.md\ndata-check.md\ncandidate-matrix.md\nfamily-brief.md\nrisk-notes.md\nraw/\n```\n\n其中：\n\n- `sources.md` 记录来源、URL、发布日期、适用年份和用途。\n- `data-check.md` 记录分数、位次、一分一段、批次线、本科线/专科线和口径风险。\n- `candidate-matrix.md` 记录候选院校/专业组、计划数、限制条件、证据链接、风险层级和待核验问题。\n- `family-brief.md` 面向学生和家长总结决策摘要、冲稳保垫分层、重大风险和电话确认清单。\n- `risk-notes.md` 记录不能直接下结论的来源冲突、缺失数据和下一步核验路径。\n\n## 验证\n\n在仓库根目录运行：\n\n```bash\npython3 scripts/gaokao_toolkit.py regression --cases tests/regression-cases.json\npython3 scripts/gaokao_toolkit.py validate-package --dir tests/fixtures/sample-package --province 广东 --year 2026 --category 物理类\npython3 scripts/gaokao_toolkit.py parse-table --input tests/fixtures/guangdong-score-table.md --out /tmp/gaokao-score-table.json --kind score --province 广东 --year 2026 --category 物理类\npython3 -m py_compile scripts/gaokao_toolkit.py\n```\n\n预期：命令返回 `ok: true` 或正常退出。\n\n## 注意\n\n`data/official-source-index.json` 是官方入口种子，不等于当年正式数据。正式研究必须进入当年省考试院发布页、阳光高考/阳光志愿、高校招生章程或其他官方材料做最终核验。\n","readmeFilename":"README.md","_rev":"1-ce48aa028220fb1e647fdc415adbc0f3"}