rocketmatter-mcp
Enables interaction with Rocketmatter legal practice management via natural language, covering matters, clients, tasks, time, invoices, calendar, documents, and trust accounting.
README
rocketmatter-mcp
MCP server for Rocketmatter — full API coverage for legal practice management. Use Rocketmatter from Claude Desktop with natural language.
What you can do
- Matters — create, update, search, manage billing info, budgets, status, custom fields, court rules
- Clients & Contacts — full CRUD, search, custom fields, contact data
- Tasks — create, assign, complete, filter by matter or user
- Time & Expenses — log billable time and expenses, manage activity types, LEDES codes
- Timers — start, pause, bill running timers
- Invoices — generate invoices, record payments, process refunds, manage templates
- Calendar — appointments linked to matters, availability checks, date range queries
- Documents — manage document records, folders, versions, download keys, templates
- Trust Accounting — view trust balances per matter
- Rates — custom matter rates, tax rates, discounts, surcharges, interest rates
- Workflow — matter workflow statuses, apply transitions
- Reports — run and retrieve firm reports
- Search — global search across all entities
- Internal Messaging — direct messages, channels
- Court Rules — apply court rules, calculate deadlines
- Matter Templates — create matters from templates
- Recurring Billing — payment plans
Requirements
- Python 3.10+
- Claude Desktop (or any MCP-compatible client)
- Rocketmatter account at app.rocketmatter.net
Installation
pip install rocketmatter-mcp
Setup
rocketmatter-mcp-setup
This opens your browser to authorize via Rocketmatter OAuth. Log in, click Allow, and the script captures the callback and saves tokens automatically. No manual credential entry required.
Verify:
rocketmatter-mcp-verify
Claude Desktop Configuration
{
"mcpServers": {
"rocketmatter": {
"command": "rocketmatter-mcp"
}
}
}
Credential storage
By default credentials are stored in your operating system's native secret store
via the cross-platform keyring library:
| OS | Backend |
|---|---|
| macOS | Keychain |
| Windows | Credential Manager |
| Linux | Secret Service (GNOME Keyring / KWallet) |
Secrets are saved under the service name rocketmatter-mcp. Nothing is written to
disk in clear text.
File fallback. On a host with no keyring backend (e.g. a headless Linux box
without Secret Service), or if you set ROCKETMATTER_MCP_USE_KEYRING=0, credentials
fall back to a ~/.rocketmatter-mcp/.env file with 0600 permissions.
Read order. Credentials resolve in the order OS keyring → process environment
→ .env file. So a rotated secret in the keyring always wins, and a
ROCKETMATTER_API_KEY exported in your shell overrides the file fallback without
touching the keyring.
Authentication Notes
Rocketmatter uses the LCS API (API key + username/password). Tokens are stored at
~/.rocketmatter-mcp/tokens.json (chmod 600) and refreshed automatically. Access
tokens expire in ~5 hours; the MCP handles refresh silently. If the token expires,
re-run rocketmatter-mcp-setup.
Example usage in Claude
"Search my matters for Smith"
"Log 2 hours on matter 456 for drafting the complaint"
"Get all upcoming calendar events for matter 789"
"Run an invoice for matter 123"
"Show me the trust account balance for matter 456"
"Calculate court deadlines for rule 12 from 2026-06-01"
License
MIT
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。
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。
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。
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。
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。
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。
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。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。