Lodestar MCP Server
启动项目
uv venvuv syncmcp dev server.py每次修改代码后,都需要重新运行
mcp dev server.py注意:你应该研究如何进行热重载
查看其他示例 mcp 服务器如何运行及其各自的工具暴露
可能的后续步骤:
我应该将 API 密钥和其他内容公开为资源吗?还是应该将平台 URL 作为资源?
如果用户没有当前用户的 api_key 和 project_id,其中一个工具应该为用户获取新的 api_key 和 project_id,但我需要以某种方式获取正确的用户
看看其他工具是如何做的
编写 mcp 提示,帮助 llms 了解如何编写提示以获得最佳结果
要查看的服务器:
看起来其中大多数都在使用 ts,也许我也应该换个方式
实际上,我不明白为什么 TypeScript 会更好,特别是在我们的用例中,所以坚持使用可读性更好的 Python 可能会更好
潜在贡献
开发人员向 MCP 编写代码时的热重载
Available Tools
1 tooldoc_queryC
Query project documentation.
Args:
query (str): The query text from the user
api_key (str): Authentication key for the request
project_id (str): Identifier for the target project
Returns:
ContextResponse: The generated response
Inputs: query (required), api_key (required), project_id (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 1 tool
With only one tool, there is no possibility of ambiguity or overlap between tools. The single tool 'doc_query' has a clearly defined purpose that cannot be confused with any other tool in the set.
The single tool name 'doc_query' follows a clear verb_noun pattern. With only one tool, the naming is inherently consistent as there are no other tools to compare against or create inconsistencies with.
A single tool feels thin for a documentation query server, suggesting limited functionality. While it might be appropriate for a minimal prototype, a production documentation server would typically offer multiple operations like search, browse, filter, or manage documentation, making this count borderline inadequate.
The tool surface is severely incomplete for a documentation domain. While 'doc_query' covers querying, there are obvious gaps: no tools for listing available documentation, managing documentation sets, filtering by categories, or handling documentation updates. This will likely cause agent failures when trying to perform comprehensive documentation tasks.
Related MCP Connectors
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
MCP server for generating rough-draft project plans from natural-language prompts.
MCP server for building and testing AI agents with multi-model experimentation and insights.
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceA production-ready MCP server built with FastAPI, providing an enhanced tool registry for creating, managing, and documenting AI tools for Large Language Models (LLMs).34-
- FlicenseNot gradedqualityDmaintenanceA template/boilerplate MCP server for building custom tools and integrations that enable seamless connections between AI applications and external data sources.-
- AlicenseNot gradedqualityDmaintenanceAn MCP server that allows users to define custom tools via a configuration file for tasks such as retrieving project architecture and managing task lists. It also includes a plugin for generating structured and optimized AI prompts based on specific context sections and instructions.2MIT
- FlicenseNot gradedqualityCmaintenanceA Python MCP server providing AI agents with tools for local knowledge base search, business context retrieval, and text summarization prompt generation.-