deeptempo-mcp-servers
Provides tools for DeepTempo AI SOC including findings and case management, investigation workflow orchestration, action approval workflows, and MITRE ATT&CK layer generation.
README
DeepTempo MCP Servers
Model Context Protocol (MCP) servers for DeepTempo AI SOC.
Overview
This package provides 4 MCP servers for DeepTempo AI SOC:
- deeptempo-findings - Findings and case management (FastMCP)
- tempo-flow - Investigation workflow orchestration
- approval - Action approval workflow
- attack-layer - MITRE ATT&CK layer generation
Installation
# Install from source (development)
pip install -e .
# Install deeptempo-core dependency first
pip install -e ../deeptempo-core
# Install from git
pip install git+https://github.com/YOUR_USERNAME/deeptempo-mcp-servers.git
Usage
Running Individual Servers
# deeptempo-findings server (FastMCP)
python servers/deeptempo_findings.py
# tempo-flow server
python servers/tempo_flow.py
# approval server
python servers/approval.py
# attack-layer server
python servers/attack_layer.py
Using with Claude Desktop
Add to your Claude Desktop MCP configuration (~/.config/claude/mcp.json):
{
"mcpServers": {
"deeptempo-findings": {
"command": "python3",
"args": ["/path/to/deeptempo-mcp-servers/servers/deeptempo_findings.py"],
"cwd": "/path/to/your/project"
},
"tempo-flow": {
"command": "python3",
"args": ["/path/to/deeptempo-mcp-servers/servers/tempo_flow.py"],
"cwd": "/path/to/your/project"
},
"approval": {
"command": "python3",
"args": ["/path/to/deeptempo-mcp-servers/servers/approval.py"],
"cwd": "/path/to/your/project"
},
"attack-layer": {
"command": "python3",
"args": ["/path/to/deeptempo-mcp-servers/servers/attack_layer.py"],
"cwd": "/path/to/your/project"
}
}
}
Server Descriptions
deeptempo-findings
Provides comprehensive access to findings and cases:
Tools:
list_findings- List security findings with filtersget_finding- Get detailed finding informationlist_cases- List investigation casesget_case- Get detailed case informationcreate_case- Create new investigation caseupdate_case- Update case detailsadd_finding_to_case- Link finding to case- And 17+ more tools for case management
tempo-flow
Investigation workflow orchestration:
Tools:
tempo_get_workflows- List available workflowstempo_run_workflow- Execute investigation workflow
approval
Action approval workflow:
Tools:
create_approval_action- Submit action for approvallist_approval_actions- List pending/approved actionsget_approval_action- Get action detailsapprove_action- Approve pending actionreject_action- Reject pending action
attack-layer
MITRE ATT&CK layer generation:
Tools:
get_attack_layer- Get ATT&CK Navigator layerget_technique_rollup- Get technique statisticsget_findings_by_technique- Find findings for techniquecreate_attack_layer- Create custom layer
Configuration
The servers use configuration from deeptempo-core. Set environment variables:
# Database
export DATABASE_URL=postgresql://user:pass@localhost:5432/deeptempo_soc
# Or individual components
export POSTGRES_HOST=localhost
export POSTGRES_PORT=5432
export POSTGRES_DB=deeptempo_soc
export POSTGRES_USER=deeptempo
export POSTGRES_PASSWORD=your_password
# Demo mode (uses generated sample data)
export DEMO_MODE=true
Development
# Install with dev dependencies
pip install -e ".[dev]"
# Run tests
pytest
# Format code
black servers/
ruff check servers/
Architecture
deeptempo-mcp-servers/
├── servers/
│ ├── __init__.py
│ ├── deeptempo_findings.py # FastMCP server
│ ├── tempo_flow.py # Standard MCP server
│ ├── approval.py # Standard MCP server
│ ├── attack_layer.py # Standard MCP server
│ └── base.py # Shared utilities
└── tests/
└── test_servers.py
License
MIT License - see LICENSE file for details.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。