china-law-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_statutesA | 按自然语言问题检索中国现行法律条文。 支持口语提问（如「同事借我钱不还怎么办」「外卖吃出异物能退吗」）， 也会自动扩展为法言法语；支持「《民法典》第1254条」这类精确引用（直查优先）。 Args: query: 自然语言问题或关键词。 top_k: 返回条数，默认 6。 law: 可选，限定在某部法律内检索，支持简称如「民法典」「消保法」。 Returns: 条文列表，含 law / article_no / chapter / text / score / via。 |
| get_articleA | 按法律名与条号取条文原文。 Args: law: 法律名，支持简称，如「民法典」「中华人民共和国刑法」。 article_no: 条号，中文或阿拉伯数字均可，如「1254」「第一千二百五十四条」。 Returns: 找到则返回条文（含所属编章、部门法、时效状态），否则 None。 |
| list_lawsA | 浏览库内法律目录（可按名称关键词或部门法筛选）。 Args: keyword: 法律名包含的关键词，如「劳动」「行政」。 department: 部门法，如「民法商法」「行政法」。 limit: 最多返回条数，默认 30。 |
| verify_citationA | 核验一条法律引用是否真实存在（防幻觉）。 在引用《某法》第 N 条之前调用本工具，可以避免编造或引用已废止条文。 Args: law: 法律名，支持简称，如「民法典」。 article_no: 条号，如「1254」或「第一千二百五十四条」。 Returns: verified(bool) + 条文原文（存在时）/ 原因（不存在时）。 |
| check_citations_in_textA | 抽取一段文本中的所有《法律》第 N 条引用并逐条核验，返回不存在的引用清单。 把模型生成的答案整段传进来，即可发现其中编造的引用。 Args: text: 待检查的文本（例如模型输出的一段法律分析）。 Returns: total / valid / invalid 计数，以及逐条核验结果。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
search_statutes (NL retrieval) and get_article (exact law+article lookup) overlap somewhat, especially since search_statutes also does '直查优先' for exact citations. verify_citation also retrieves a law+article, but its validation-focused return (verified bool) distinguishes it from get_article's text retrieval, and check_citations_in_text is clearly distinct as a batch extractor.
All five tools use a consistent snake_case verb_noun pattern: search_statutes, get_article, list_laws, verify_citation, check_citations_in_text. The longer batch tool name remains readable and follows the same convention.
Five tools is well-scoped for a legal retrieval/verification server, with each tool covering a distinct capability (search, exact fetch, browse, single verify, bulk verify). No redundant or filler tools.
The surface covers the core read-only legal workflow: search by NL, fetch by citation, browse the catalog, and verify single or bulk citations. Minor gaps like retrieving a full law's chapter structure or browsing by effective date exist, but the primary agent workflows are supported.