huly-mcp
MCP server for Huly project management that enables AI assistants to manage issues, track time, update statuses, and more directly from your IDE.
README
huly-mcp
MCP server for Huly project management platform. Lets AI assistants (Claude, etc.) manage issues, track time, update statuses, and more — directly from your IDE.
Tools
| Tool | Description |
|---|---|
list_projects |
List all projects in the workspace |
list_issues |
List issues with filters (project, status, assignee, priority) |
get_issue |
Get issue details with full markdown description |
update_status |
Change issue status |
update_assignee |
Change issue assignee |
add_time |
Log time spent on an issue |
add_comment |
Add a comment to an issue |
list_statuses |
List available issue statuses for a project |
list_members |
List workspace members |
time_report |
Get time report for a member for a date or date range |
set_custom_field |
Set a custom field value on an issue (requires config) |
Setup
1. Clone and install
git clone https://github.com/YarikYar/huly-mcp.git
cd huly-mcp
npm install
2. Configure credentials
Copy the example env file and fill in your Huly credentials:
cp .env.example .env
HULY_URL=https://huly.app
HULY_EMAIL=your@email.com
HULY_PASSWORD=your_password
HULY_WORKSPACE=your-workspace-id
To find your workspace ID, go to Huly settings or check the URL when logged in.
3. Add to Claude Code
Add to your .mcp.json (project-level or ~/.claude/.mcp.json for global):
{
"mcpServers": {
"huly": {
"command": "npx",
"args": ["tsx", "/path/to/huly-mcp/src/server.ts"],
"env": {
"HULY_URL": "https://huly.app",
"HULY_EMAIL": "your@email.com",
"HULY_PASSWORD": "your_password",
"HULY_WORKSPACE": "your-workspace-id"
}
}
}
}
Or use .env file and omit the env block:
{
"mcpServers": {
"huly": {
"command": "npx",
"args": ["tsx", "/path/to/huly-mcp/src/server.ts"]
}
}
}
4. Add to Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"huly": {
"command": "npx",
"args": ["tsx", "/path/to/huly-mcp/src/server.ts"],
"env": {
"HULY_URL": "https://huly.app",
"HULY_EMAIL": "your@email.com",
"HULY_PASSWORD": "your_password",
"HULY_WORKSPACE": "your-workspace-id"
}
}
}
}
Custom fields
Huly stores custom fields as mixins with auto-generated IDs. To use set_custom_field, you need to find the field ID and map it to a friendly name.
Finding custom field IDs
Create a temporary script dump-fields.ts:
import dotenv from 'dotenv'
import path from 'path'
dotenv.config({ path: path.resolve(__dirname, '.env') })
import { ConnectOptions, NodeWebSocketFactory, connect } from '@hcengineering/api-client'
import tracker from '@hcengineering/tracker'
async function main() {
const client = await connect(process.env.HULY_URL ?? 'https://huly.app', {
email: process.env.HULY_EMAIL!,
password: process.env.HULY_PASSWORD!,
workspace: process.env.HULY_WORKSPACE!,
socketFactory: NodeWebSocketFactory,
connectionTimeout: 30000
})
try {
// Pick any issue that has your custom fields filled in
const issue = await client.findOne(tracker.class.Issue, { identifier: 'DEV-1' })
if (!issue) throw new Error('Issue not found')
const mixinData = (issue as any)['tracker:mixin:IssueTypeData']
if (mixinData) {
console.log('Custom fields on this issue:')
for (const [key, value] of Object.entries(mixinData)) {
console.log(` ${key} = "${value}"`)
}
} else {
console.log('No custom fields found on this issue')
}
} finally {
await client.close()
}
}
main().catch(console.error)
Run it:
npx tsx dump-fields.ts
Output example:
Custom fields on this issue:
custom696fc1e93e1982f72ab4b92f = "main"
Configuring custom fields
Add the mapping to your env (in .env or .mcp.json):
# Format: friendlyName=hulyFieldId,anotherName=anotherId
HULY_CUSTOM_FIELDS=gitlabBranch=custom696fc1e93e1982f72ab4b92f
The set_custom_field tool will only appear if HULY_CUSTOM_FIELDS is configured. Custom field values also appear in get_issue output automatically.
Usage examples
Once connected, you can ask your AI assistant things like:
- "Show my Todo tasks in the DEV project"
- "What's the description of DEV-592?"
- "Set DEV-592 to In Progress"
- "Log 2 hours on DEV-599"
- "Assign DEV-597 to Arsenii"
- "Add a comment to DEV-592: Started working on the bonus system"
- "Set gitlabBranch on DEV-592 to @user/feat/my-feature"
- "Show my time report for yesterday"
Priority values
| Value | Meaning |
|---|---|
| 0 | No priority |
| 1 | Urgent |
| 2 | High |
| 3 | Medium |
| 4 | Low |
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。
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。