DevCycle MCP server
2026년 11월 9일 커밋AI 코딩 어시스턴트가 에디터에서 직접 DevCycle의 feature flag, 변수, variation, targeting rule을 생성, 업데이트 및 관리할 수 있습니다.
다른 이름 DevCycleHQ/cli, cli
- 스타
- 20
- Tools
- —
- 다운로드, 30일
- —
- 최신 버전
- 6.6.1
- 마지막 커밋
- 2026년 11월 9일
- 라이선스
- MIT
DevCycle MCP란?
DevCycle은(는) MCP server입니다. AI 코딩 어시스턴트가 에디터에서 직접 DevCycle의 feature flag, 변수, variation, targeting rule을 생성, 업데이트 및 관리할 수 있습니다.
프로젝트 README에서 생성된 요약
DevCycle MCP server 정보
이 MCP server는 Cursor 또는 Claude를 사용하는 개발자가 코딩 환경에서 직접 feature flag와 targeting rule을 관리할 수 있도록 지원합니다. 호스팅된 DevCycle server에 연결하면 AI 어시스턴트가 로컬 설치 없이도 전체 feature management 라이프사이클을 생성, 업데이트 및 감독할 수 있는 권한을 갖게 됩니다. 설정은 에디터 설정에서 간단한 구성으로 처리되며, 첫 사용 시 보안 인증을 위해 OAuth를 사용합니다.
이 통합을 통해 AI는 feature, 변수, variation, targeting rule을 포함한 복잡한 DevCycle 구성 요소와 상호작용할 수 있습니다. 단순한 flag 관리를 넘어, 코드베이스 내 변수 사용 감지, 타입 안정성을 위한 TypeScript 타입 정의 생성, self-targeting override 관리와 같은 고급 개발 워크플로우를 지원합니다. AI 어시스턴트와 DevCycle 사이의 이러한 연결을 통해 feature 출시 및 구성 변경을 기존 개발 워크플로우 내에서 직접 수행할 수 있습니다.
활용 사례
- 1자동화된 feature flag 관리AI 어시스턴트에게 에디터에서 직접 feature flag와 targeting rule을 생성하거나 수정하도록 요청합니다.
- 2코드베이스 변수 감사현재 소스 코드에 존재하는 모든 DevCycle 변수 사용 목록을 요청합니다.
- 3타입 안정성이 보장된 flag 구현AI 어시스턴트가 TypeScript 정의를 생성하여 feature flag 사용 시 타입 안정성을 확보하도록 합니다.
- 4override를 활용한 신속한 테스트빠른 테스트를 위해 self-targeting override를 사용하여 변수 값을 변경하도록 AI에게 지시합니다.
프로젝트의 README를 바탕으로 작성되었습니다. 모든 활용 사례는 해당 내용의 인용문과 대조하여 확인합니다.
DevCycle MCP 설치 방법
형식 확인: 2026년 11월 10일Cursor에 추가Goose에 추가Claude Code에 DevCycle MCP 설치
$ claude mcp add --transport http cli https://mcp.devcycle.com/mcpCursor에 DevCycle MCP 설치
~/.cursor/mcp.json
{
"mcpServers": {
"cli": {
"url": "https://mcp.devcycle.com/mcp"
}
}
}VS Code에 DevCycle MCP 설치
.vscode/mcp.json
{
"servers": {
"cli": {
"type": "http",
"url": "https://mcp.devcycle.com/mcp"
}
}
}VS Code CLI에 DevCycle MCP 설치
$ code --add-mcp '{"name":"cli","type":"http","url":"https://mcp.devcycle.com/mcp"}'Codex에 DevCycle MCP 설치
$ codex mcp add cli --url https://mcp.devcycle.com/mcpCodex config.toml에 DevCycle MCP 설치
~/.codex/config.toml
[mcp_servers.cli] url = "https://mcp.devcycle.com/mcp"
Gemini CLI에 DevCycle MCP 설치
$ gemini mcp add -s user --transport http cli https://mcp.devcycle.com/mcpGemini settings.json에 DevCycle MCP 설치
~/.gemini/settings.json
{
"mcpServers": {
"cli": {
"url": "https://mcp.devcycle.com/mcp",
"type": "http"
}
}
}자동 검사
2026년 11월 13일에 실행됨전체 보고서 →아직 인증되지 않았습니다. 서버는 80점 이상의 점수, 심각한 결함 없음, 그리고 검사에 충분한 데이터를 갖추어야 합니다. 검사 방식 · 점수를 높이기 위해 수정할 항목
- 소스 저장소 연결 및 접근 가능devcyclehq/cli is public and reachablegithub.com/devcyclehq/cli ↗통과
- 패키지가 이 저장소를 소스로 명시검사되지 않음
- 빌드 출처(Sigstore 증명)검사되지 않음
- 오픈 소스 라이선스 (OSI)MITspdx.org/licenses/MIT.html ↗통과
- 최근 1년 내 커밋 또는 릴리스Last commit or release 2026-11-09github.com/devcyclehq/cli ↗통과
- OpenSSF Scorecard검사되지 않음
- 패키지 이름이 인기 패키지와 유사하지 않음검사되지 않음
- 패키지 및 해당 종속성에 중요한 보안 권고 없음검사되지 않음
- 모델에 대한 숨겨진 지침 없음0 tool descriptions and the README read; no override phrases, credential paths or hidden characters통과
- 파괴적인 도구에 라벨 표시됨검사되지 않음
- 원격 엔드포인트가 MCP 핸드셰이크에 응답HTTPS endpoint answered 401 (sign-in required)통과
README
GitHub에서 보기DevCycle CLI & MCP Server
This repository contains the DevCycle CLI for managing feature flags from the command line, plus an MCP (Model Context Protocol) server that enables AI coding assistants to interact with DevCycle.
Major features include:
- Fully manage your Features, Variables, Variations and Targeting Rules from the command line
- Detect and list DevCycle Variable usages in your codebase
- Manage your Self-Targeting Overrides to quickly switch between Variable values
- Generate type definitions for type-safe usage of DevCycle (Typescript only)
- MCP (Model Context Protocol) server for AI-powered feature flag management with Cursor and Claude
The CLI can be customized in several ways using command-line args or by creating a configuration file.
- Command Topics
- Pattern captures the key including surrounding quotes
- Pattern strips quotes, capturing only the content
MCP Server for AI Assistants
The DevCycle MCP (Model Context Protocol) server enables AI coding assistants like Cursor and Claude to manage feature flags directly from your development environment. DevCycle offers a hosted MCP server that requires no local installation.
Quick Setup (No Installation Required)
-
Configure your AI assistant to use the hosted MCP server:
-
Cursor: Add to
.cursor/mcp_settings.json:{ "mcpServers": { "devcycle": { "url": "https://mcp.devcycle.com/mcp" } } } -
Claude Desktop: Add to your Claude config file:
{ "mcpServers": { "devcycle": { "command": "npx", "args": [ "mcp-remote", "https://mcp.devcycle.com/mcp" ] } } }
-
-
That's it! The server will guide you through OAuth authentication when you first use it.
Your AI assistant can now create, update, and manage feature flags on your behalf.
For local installation options, detailed configuration, available tools, and advanced usage, see the complete MCP documentation.
CLI Documentation
Setup
Install the CLI
Using NPM
$ npm install -g @devcycle/cli
Or alternatively, using homebrew
$ brew tap devcyclehq/cli
$ brew install devcycle
Authentication
Many of the CLI commands require DevCycle API authorization. There are several ways to provide these credentials.
Using Access Tokens (preferred)
Login Command
By using the login sso command, the CLI will retrieve and store an access token, which is valid for 24 hours.
The login again command can be used to retrieve a new access token using the saved project and organization without prompting for them.
This process will open browser windows to interact with the DevCycle universal login page. It will first obtain a personal access token, then prompt you to choose an organization. A second browser window is used to authenticate the CLI with your chosen organization.
To switch organizations once logged in, the organizations select command can be used.
If executing the CLI in a containerized environment, please ensure one of the following PORTs can be accessed via Port Forwarding: 2194 (default), 2195, 2196 or 8080. This will allow the authentication process to complete and set the access token appropriately.
Repo Init Command
The repo init command behaves in the same way as login sso, but creates a repo configuration file and stores the project and organization choices there instead.
Using Client Credentials
Client Credentials in Auth File
Use the dvc status command to find the configuration file location for your platform. The credentials can be stored in the file pointed to by the Auth config path. Create the file if it does not exist, with the following contents.
clientCredentials:
client_id: <your client id>
client_secret: <your client secret>
This file should not be checked in to version control.
The default location is based on the oclif configDir
If you intend to run the CLI using options that override config file locations, the dvc status command command can be run with those options to confirm that the file locations are as expected.
Project Selection
You also need to specify the default project ID for the CLI to use.
If there is a repo configuration file, the dvc diff and dvc usages commands will use the project defined there.
Otherwise, this is chosen during login or set using the project select command
Environment Variables
Set the following environment variables:
$ export DEVCYCLE_CLIENT_ID=<your client id>
$ export DEVCYCLE_CLIENT_SECRET=<your client secret>
$ export DEVCYCLE_PROJECT_KEY=<your project key>
Command-Line Arguments
The CLI can be run with the following arguments:
$ dvc --client-id=<your client id> --client-secret=<your client secret> --project=<your project key>
Github Action
The Devcycle Github actions are configured with auth information through the project-key, client-id and client-secret configuration parameters. This is passed to the CLI via command line arguments.
Usage
$ npm install -g @devcycle/cli
$ dvc COMMAND
running command...
$ dvc (--version)
@devcycle/cli/6.6.1 linux-x64 node-v24.21.0
$ dvc --help [COMMAND]
USAGE
$ dvc COMMAND
...
Command Topics
dvc alias- Manage repository variable aliases.dvc autocomplete- display autocomplete installation instructionsdvc cleanup- Replace a DevCycle variable with a static value in the current version of your code. Currently only JavaScript is supported.dvc diff- Print a diff of DevCycle variable usage between two versions of your code.dvc environments- Create a new Environment for an existing Feature.dvc features- Create, view, or modify Features with the Management API.dvc generate- Generate Devcycle related files.dvc help- Display help for dvc.dvc identity- View or manage your DevCycle Identity.dvc keys- Retrieve SDK keys from the Management API.dvc login- Log in to DevCycle.dvc logout- Discards any auth configuration that has been stored in the auth configuration file.dvc organizations- List or switch organizations.dvc overrides- Create, view, or modify Overrides for a Project with the Management API.dvc projects- Create, or view Projects with the Management API.dvc repo- Manage repository configuration.dvc status- Check CLI status.dvc targeting- Create, view, or modify Targeting Rules for a Feature with the Management API.dvc usages- Print all DevCycle variable usages in the current version of your code.dvc variables- Create, view, or modify Variables with the Management API.dvc variations- Create a new Variation for an existing Feature.
Repo Configuration
The following commands can only be run from the root of a configured repository
Many of the options available as command-line args can also be specified using a repo configuration file. The default
location for this file is <REPO ROOT>/.devcycle/config.yml.
This location can be overridden using the --repo-config-path flag.
The configuration file format is documented below:
## the project and organization to use when connecting to the DevCycle Rest API for this repo
project: "project-key"
org:
id: "org_xxxxxx"
name: "unique-org-key"
display_name: "Human Readable Org Name"
## block for configuring "code insights" features like diff and variable usage scanning
## use this section to improve the detection of DevCycle usage within your code
codeInsights:
## add additional names to check for when looking for instances of DVCClient from an SDK
clientNames:
- "dvcClient"
## map the values used in your code to the corresponding variable key in DevCycle
variableAliases:
"VARIABLES.ENABLE_V1": "enable-v1"
## additional regex patterns used to match variables for a specific file extension
matchPatterns:
## file extension to override for, containing a list of patterns to use
js:
- dvcClient\.variable\(\s*["']([^"']*)["']
## an array of file glob patterns to include in usage scan
includeFiles:
- "*.[jt]s"
## an array of file glob patterns to exclude from usage scan
excludeFiles:
- "dist/*"
Match Patterns and Aliases
When identifying variable usages in the code, the CLI will identify DevCycle SDK methods by default. To capture other usages you may define match patterns. Match patterns are defined by file extension.
:::note Each pattern must include exactly one capture group for the variable key. Capture the entire key value (including surrounding quotes if you choose the “with quotes” pattern). :::
Match patterns can be defined in the configuration file, for example:
codeInsights:
matchPatterns:
js:
- customVariableGetter\(\s*["']([^"']*)["']
ts:
- customVariableGetter\(\s*["']([^"']*)["']
jsx:
- customVariableHook\(\s*["']([^"']*)["']
- customVariableGetter\(\s*["']([^"']*)["']
tsx:
- customVariableHook\(\s*["']([^"']*)["']
- customVariableGetter\(\s*["']([^"']*)["']
Capturing with or without quotes
Match patterns can capture variable keys with or without quotes, which affects whether aliases are needed:
With quotes:
# Pattern captures the key including surrounding quotes
- dvcClient\.variable\(\s*(["'][^"']*["'])\s*,
- Matches:
dvcClient.variable('my-variable', default) - Captures:
'my-variable'(with quotes)
Without quotes (aliases or generated constants required):
# Pattern strips quotes, capturing only the content
- dvcClient\.variable\(\s*["']([^"']*)["']
- Matches:
dvcClient.variable('my-variable', default) - Captures:
my-variable
Match patterns can also be passed directly to relevant commands using the --match-pattern flag:
dvc usages --match-pattern ts="customVariableGetter\(\s*["']([^"']*)["']" js="customVariableGetter\(\s*["']([^"']*)["']"
When testing your regex the --show-regex flag can be helpful. This will print all patterns used to find matches in your codebase.
dvc usages --show-regex
Custom Wrapper Functions
If you use wrapper functions around the SDK, add patterns for them.
Example: Custom wrapper functions
codeInsights:
matchPatterns:
ts:
# Matches: getFeatureFlag('my-variable', defaultValue)
- getFeatureFlag\(\s*([^,)]*)\s*,
# Matches: isFeatureEnabled('my-variable')
- isFeatureEnabled\(\s*([^,)]*)\s*\)
tsx:
# Matches: useFeatureFlag('my-variable', defaultValue)
- useFeatureFlag\(\s*([^,)]*)\s*,
DevCycle MCP 질문
DevCycle MCP를 Claude Code에 추가하려면 어떻게 하나요?
터미널에서 claude mcp add --transport http cli https://mcp.devcycle.com/mcp를 실행합니다. 모든 프로젝트에서 사용할 수 있게 하려면 --scope user를 추가합니다.
DevCycle MCP를 Cursor에 추가하려면 어떻게 하나요?
~/.cursor/mcp.json를 열고 mcpServers 아래에 서버를 추가합니다. 위 설치 섹션에 표시된 Cursor 구성을 사용합니다.
DevCycle MCP server에 OAuth 인증이 필요한가요?
네, 호스팅된 서버를 처음 사용할 때 OAuth 인증 과정을 안내합니다.
MCP를 통해 AI 어시스턴트가 DevCycle feature flag로 무엇을 할 수 있나요?
feature, 변수, variation, targeting rule을 포함하여 사용자를 대신해 feature flag를 생성, 업데이트 및 관리할 수 있습니다.
어떤 AI 어시스턴트가 DevCycle MCP server와 호환되나요?
Cursor 및 Claude와 같은 AI 코딩 어시스턴트를 위해 설계되었으며, 두 도구 모두를 위한 설정 스니펫을 제공합니다.