Claude Team MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GEMINI_API_KEY | No | Gemini API Key for fast tasks (optional) | |
| OPENAI_API_KEY | No | OpenAI API Key for balanced tasks (optional) | |
| ANTHROPIC_API_KEY | No | Anthropic API Key for powerful tasks (optional) | |
| CLAUDE_TEAM_MAIN_KEY | Yes | 主模型 API Key (required) | |
| CLAUDE_TEAM_MAIN_URL | No | 主模型 API 地址 (optional) | |
| CLAUDE_TEAM_MAIN_MODEL | No | 主模型 ID (optional, default: gpt-4o) | gpt-4o |
| CLAUDE_TEAM_MODEL1_KEY | No | 模型1 API Key (optional, defaults to MAIN key) | |
| CLAUDE_TEAM_MODEL1_URL | No | 模型1 API 地址 (optional, defaults to MAIN URL) | |
| CLAUDE_TEAM_MODEL2_KEY | No | 模型2 API Key (optional, defaults to MAIN key) | |
| CLAUDE_TEAM_MODEL2_URL | No | 模型2 API 地址 (optional, defaults to MAIN URL) | |
| CLAUDE_TEAM_MODEL3_KEY | No | 模型3 API Key (optional, defaults to MAIN key) | |
| CLAUDE_TEAM_MODEL3_URL | No | 模型3 API 地址 (optional, defaults to MAIN URL) | |
| CLAUDE_TEAM_MODEL1_NAME | No | 模型1 ID (optional) | |
| CLAUDE_TEAM_MODEL2_NAME | No | 模型2 ID (optional) | |
| CLAUDE_TEAM_MODEL3_NAME | No | 模型3 ID (optional) | |
| CLAUDE_TEAM_MAIN_PROVIDER | No | 主模型提供商 (openai | anthropic | gemini) | |
| CLAUDE_TEAM_MODEL1_PROVIDER | No | 模型1提供商 (openai | anthropic | gemini) | |
| CLAUDE_TEAM_MODEL2_PROVIDER | No | 模型2提供商 (openai | anthropic | gemini) | |
| CLAUDE_TEAM_MODEL3_PROVIDER | No | 模型3提供商 (openai | anthropic | gemini) |
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 | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| team_workB | 让 AI 开发团队协作完成任务。团队包含前端专家、后端专家、QA专家，会智能分配任务并互相协作。 |
| ask_expertB | 向特定专家咨询问题 |
| code_reviewC | 让专家审查代码 |
| fix_bugD | 让 QA 专家修复 Bug |
| history_listC | 查看团队协作历史记录列表 |
| history_getB | 获取某次协作的详细记录 |
| history_searchC | 搜索协作历史记录 |
| history_contextB | 获取最近的协作上下文，可用于继续之前的工作 |
| usage_statsA | 查看各模型的使用统计（调用次数、成功率、平均耗时） |
| team_dashboardA | 查看团队当前状态：可用专家、模型配置、最近活动 |
| cost_estimateB | 预估任务执行成本（Token 用量、预计耗时） |
| explain_planB | 解释 Tech Lead 会如何分配任务（不实际执行） |
| read_project_filesC | 读取项目文件内容，让专家了解代码上下文 |
| generate_commit_messageC | 根据代码变更生成 Git commit message |
| analyze_project_structureB | 分析项目结构，识别技术栈和架构 |
| list_workflowsA | 列出所有可用的工作流模板 |
| run_workflowB | 使用指定工作流执行任务 |
| suggest_workflowB | 根据任务自动推荐合适的工作流 |
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 18 tools
Most tools have distinct purposes (e.g., analyze_project_structure vs. team_work), but some overlap exists between 'ask_expert', 'code_review', and 'fix_bug' as they all involve expert interaction, though at different stages of development.
Tools predominantly use snake_case with descriptive names (e.g., 'analyze_project_structure', 'generate_commit_message'), but some like 'history_context' and 'team_dashboard' are more noun-centric, breaking the verb_noun pattern slightly.
At 18 tools, the server is slightly above the ideal range of 3-15 but still manageable. The count aligns with the broad scope of team collaboration, workflow management, and history tracking.
The tool set covers key operations for team collaboration (analysis, code review, bug fixing, workflow execution, history) but lacks administrative features like workflow creation or team member management, leaving minor gaps.