Folk CRM MCP Server
Enables AI assistants to manage Folk CRM data, including contacts, companies, notes, and reminders. It supports searching for entities, logging interactions, and setting follow-up tasks through the Folk REST API.
README
Folk CRM MCP Server
An MCP (Model Context Protocol) server that provides access to Folk CRM functionality, allowing AI assistants to manage contacts, companies, notes, reminders, and more.
Features
- Smart Search: Find people and companies by name with minimal token usage
- Two-Phase Lookup: Quick search returns IDs, then fetch full details as needed
- Contact Management: Create, update, and delete people and companies
- Notes & Reminders: Attach context to your contacts
- Interaction Logging: Track emails, meetings, and calls
Adding to Claude Code
From Registry (Published)
# Configure your Folk API key
mpak config set @nimblebraininc/folk api_key=your_api_key_here
# Add to Claude Code
claude mcp add folk -- mpak run @nimblebraininc/folk
Local Development
# Clone and enter the repo
git clone https://github.com/NimbleBrainInc/mcp-folk.git
cd mcp-folk
# Install dependencies
uv sync
# Build the bundle
make pack
# Configure your API key
mpak config set @nimblebraininc/folk api_key=your_api_key_here
# Add to Claude Code (use absolute path)
claude mcp add folk -- mpak run --local /path/to/mcp-folk/mcp-folk-0.1.0-darwin-arm64.mcpb
Configuration
Getting Your Folk API Key
- Log in to your Folk workspace
- Go to Settings > API
- Create a new API key
- Copy the key and configure with
mpak config set
Available Tools
Search (Use First)
| Tool | Purpose |
|---|---|
find_person(name) |
Find people by name, returns {found, matches: [{id, name, email}]} |
find_company(name) |
Find companies by name, returns {found, matches: [{id, name, industry}]} |
Details (After Finding)
| Tool | Purpose |
|---|---|
get_person_details(person_id) |
Full person info including all fields |
get_company_details(company_id) |
Full company info including all fields |
Browse
| Tool | Purpose |
|---|---|
browse_people(page, per_page) |
Paginated list of all people |
browse_companies(page, per_page) |
Paginated list of all companies |
Actions
| Tool | Purpose |
|---|---|
add_person(first_name, ...) |
Create new person |
add_company(name, ...) |
Create new company |
update_person(person_id, ...) |
Update person fields |
update_company(company_id, ...) |
Update company fields |
delete_person(person_id) |
Delete a person |
delete_company(company_id) |
Delete a company |
Notes & Reminders
| Tool | Purpose |
|---|---|
add_note(person_id, content) |
Add note to person |
get_notes(person_id) |
Get notes for person |
set_reminder(person_id, reminder, when) |
Set a reminder |
log_interaction(person_id, type, when) |
Log an interaction |
Utility
| Tool | Purpose |
|---|---|
whoami() |
Get current authenticated user |
Common Use Cases
Look up contacts
- "Is Sarah Chen in my CRM?"
- "Find everyone at Acme Corp"
- "What's John's email?"
Add contacts after meetings
- "Add Mike Johnson from today's meeting, he's a PM at Stripe"
- "Create a contact for lisa@example.com"
Take notes
- "Add a note to Sarah: discussed Q2 roadmap, she's interested in enterprise plan"
- "What are my notes on the Acme deal?"
Set follow-ups
- "Remind me to follow up with John next Tuesday"
- "Set a reminder to check in with Sarah in 2 weeks"
Log interactions
- "Log that I had a call with Mike today"
- "Record my meeting with the Acme team"
Browse contacts
- "Show me my recent contacts"
- "List all companies in my CRM"
Example Flow
User: "I just had coffee with Alex Rivera, she's interested in our API. Remind me to send her docs next week."
AI: find_person("Alex Rivera")
→ {"found": true, "matches": [{"id": "abc123", "name": "Alex Rivera", "email": "alex@techco.io"}]}
AI: add_note("abc123", "Had coffee - interested in API, wants to see docs")
→ {"id": "note456", "added": true}
AI: log_interaction("abc123", "meeting", "2024-01-15T10:00:00Z")
→ {"id": "int789", "logged": true}
AI: set_reminder("abc123", "Send API docs to Alex", "2024-01-22T09:00:00Z")
→ {"id": "rem012", "set": true}
AI: "Done! I've added a note about your coffee chat, logged the meeting, and set a reminder for next Monday to send her the API docs."
Development
# Install dev dependencies
uv sync --dev
# Run tests
uv run pytest tests/ -v
# Format code
uv run ruff format .
# Lint
uv run ruff check .
# Type check
uv run mypy src/
# Run all checks
make check
# Build bundle for testing
make pack
API Reference
This server uses the Folk REST API. Key endpoints:
- Base URL:
https://api.folk.app/v1 - Authentication: Bearer token
- Rate limits apply (see Folk documentation)
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。
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。