Alibaba Cloud DevOps MCP Server

Official

Enables AI assistants to interact seamlessly with the Yunxiao platform for code, project, pipeline, package, and test management.

How to install ↓ Compare GitHub source repository ↗

Category: Dev Tools #60 of 438
Stars: ★ 175 Very popular
Transport: stdio (local process) · SSE · Streamable HTTP
Runtime: Node.js 18+ · Docker · Hosted (no local runtime)
Credentials: API key / credential required
License: Apache-2.0
Last commit: 1 day ago
Tools: 100

64FMRS · C

This server is officially provided by Alibaba Cloud, feature-rich, covering the core DevOps scenarios of the Yunxiao platform. It is easy to install, supporting stdio, SSE, and Streamable HTTP transports. Pay attention to security configuration and permission management.

Strongest · Documentation 15/20 Weakest · Reliability 10/20

Reliability

10/20

Security and permissions

12/20

Maintenance

14/20

Documentation

15/20

Setup experience

13/20

Why each score

Reliability 10/20: Source code shows MCP server implementation with tool registration and handling. However, this static review does not execute code, and the server depends on external APIs (Aliyun DevOps), so real behavior cannot be verified. There are some test directories, but no CI config evidence, so reliability score is capped at 12. Given the large number of tools and likely incomplete error handling, score 10.
Security and permissions 12/20: Server requires a personal access token (PAT) passed via environment variable or HTTP header. README instructs users to obtain a token but does not emphasize least privilege. Server provides many operations including create, delete, update, but no confirmation mechanisms for dangerous ops are mentioned. No malicious code or data exfiltration evidence. Deductions for lack of permission control and confirmation.
Maintenance 14/20: Repo is active with recent commits and releases (npm package), Apache-2.0 license. No security response channel provided. Community support via discussion group. Maintenance is active but governance gaps exist.
Documentation 15/20: README is comprehensive: installation, config, tool list, Docker usage, hosted service guide. Parameters and examples are provided. Missing troubleshooting and limitations such as rate limits, error handling. Good but improvable.
Setup experience 13/20: Multiple installation methods: npx, Docker, remote hosted. README gives clear config examples. Requires Node.js >= 18 and PAT. Static review cannot verify runtime, but steps are clear. Score capped at 15 due to lack of tests/CI evidence.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkChanges third-party account data

Best for

Development teams using the Alibaba Cloud Yunxiao platform
Teams wanting to use AI to assist DevOps workflows
Teams needing automation and intelligent code management and pipeline operations

Not for

Teams not using the Alibaba Cloud Yunxiao platform
Users needing direct integration with local Git or non-Yunxiao code repositories
Users with strict security and compliance requirements who do not want to expose code content to external AI services

Required permissions

Requires a Yunxiao personal access token (YUNXIAO_ACCESS_TOKEN) with read and write permissions to all relevant APIs.
The token grants access to organization management, project collaboration, code management, pipeline management, artifact repository management, application delivery, and testing management.

Risks and side effects

Token leakage could lead to organizational data security risks.
AI might accidentally perform destructive operations (e.g., deleting branches or files).
The tool calls Yunxiao APIs, which may incur costs or quota consumption.

Setup

Before you start

Runtime：Node.js 18+ · Docker · Hosted (no local runtime)

YUNXIAO_ACCESS_TOKEN requiredsecret Yunxiao Personal Access Token obtained from the Yunxiao console (see Alibaba Cloud help docs); must grant read/write permissions to organization, project, code, pipeline, packages, application delivery and testing APIs.

Other optional settings (7)

YUNXIAO_API_BASE_URL optional Optional API base URL for Region edition deployments, e.g. https://your-org.devops.aliyuncs.com; defaults to central station https://openapi-rdc.aliyuncs.com.

DEVOPS_TOOLSETS optional Optional comma-separated toolset names (e.g. code-management,project-management) to enable only selected tools.

MCP_TRANSPORT optional Docker transport mode: sse, streamable-http, or both; stdio is the default.

PORT optional Listen port for Docker HTTP transports (default 3000 in examples).

MCP_STREAMABLE_PATH optional Optional Streamable HTTP endpoint path (default /mcp).

MCP_HTTP_HOST optional Bind host for HTTP transports (default 0.0.0.0).

MCP_ALLOWED_HOSTS optional Optional comma-separated allowed Host header values.

Ensure Node.js version >= 18.0.0.
Obtain a Personal Access Token from the Alibaba Cloud DevOps platform, granting read and write permissions to all APIs under organization management, project collaboration, code management, pipeline management, artifact repository management, application delivery and testing management.
Add the following configuration to your MCP client config (stdio mode):

{
  "mcpServers": {
    "yunxiao": {
      "command": "npx",
      "args": ["-y", "alibabacloud-devops-mcp-server"],
      "env": {
        "YUNXIAO_ACCESS_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}

If using a Region edition, also set the environment variable YUNXIAO_API_BASE_URL to your organization's domain.
For Docker installations, refer to the Docker section in the README.

claude_desktop_config.json

{
  "mcpServers": {
    "yunxiao": {
      "command": "npx",
      "args": [
        "-y",
        "alibabacloud-devops-mcp-server"
      ],
      "env": {
        "YUNXIAO_ACCESS_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json

{
  "servers": {
    "yunxiao": {
      "command": "npx",
      "args": [
        "-y",
        "alibabacloud-devops-mcp-server"
      ],
      "env": {
        "YUNXIAO_ACCESS_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal

claude mcp add yunxiao -e 'YUNXIAO_ACCESS_TOKEN=<YOUR_TOKEN>' -- npx -y alibabacloud-devops-mcp-server

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After configuring the client, tools such as get_current_organization_Info and list_pipelines should appear in the client tool list; asking 'Which Yunxiao organizations am I in?' and getting a result confirms the connection.

Troubleshooting

Ensure the YUNXIAO_ACCESS_TOKEN environment variable is set correctly.
Check network connectivity to the Yunxiao API domain (e.g., openapi-rdc.aliyuncs.com or your organization's dedicated domain).
Verify that the token has the required permissions.
If using Docker, ensure proper container port mapping.

Things to try

Once connected, you can ask your AI assistant things like:

List the Yunxiao organizations I belong to and my role in each
Search the work items in my project and summarize open ones
Generate pipeline YAML from this description and create a pipeline
Create a branch in my Codeup repository and submit a merge request

Tools 100

get_current_organization_Info read-only

get_user_organizations read-only

get_organization_role read-only

get_organization_departments read-only

get_organization_department_info read-only

get_organization_department_ancestors read-only

get_organization_members read-only

get_organization_member_info read-only

Show 92 more tools

get_organization_member_info_by_user_id read-only

search_organization_members read-only

list_organization_roles read-only

create_branch writes

Create a branch

delete_branch destructive

Delete a branch

get_branch read-only

Get branch information

list_branches read-only

Get branch list

create_file writes

Create a file

delete_file destructive

Delete a file

get_file_blobs read-only

Get file content

list_files read-only

Query file tree

update_file writes

Update file content

create_change_request writes

Create a merge request

create_change_request_comment writes

Create a comment on a merge request

get_change_request read-only

Query merge request

list_change_request_patch_sets read-only

Query merge request version list

list_change_request read-only

Query merge request list

list_change_request_comments read-only

Query merge request comment list

get_compare read-only

Compare code

get_repository read-only

Get repository details

list_repositories read-only

Get repository list

list_commits read-only

List commits in a repository

get_commit read-only

Get commit details

create_commit_comment writes

Create a comment on a commit

get_project read-only

Get project details

search_projects read-only

Search projects

get_sprint read-only

Get sprint details

list_sprints read-only

List sprints in a project

create_sprint writes

Create a sprint

update_sprint writes

Update a sprint

search_programs read-only

Search for programs

list_program_versions read-only

List program versions

list_versions read-only

List project versions

create_version writes

Create a version

update_version writes

Update a version

delete_version destructive

Delete a version

get_work_item read-only

Get work item details

update_work_item writes

Update a work item

search_workitems read-only

Search work items

get_work_item_types read-only

Get work item types

create_work_item writes

Create work item

list_all_work_item_types read-only

List all work item types in an organization

list_work_item_types read-only

List work item types in a project space

get_work_item_type read-only

Get details of a specific work item type

list_work_item_relation_work_item_types read-only

List work item types that can be related to a specific work item

get_work_item_type_field_config read-only

Get field configuration for a specific work item type

get_work_item_workflow read-only

Get workflow information for a specific work item type

list_work_item_comments read-only

List comments for a specific work item

create_work_item_comment writes

Create a comment for a specific work item

list_workitem_attachments read-only

List attachments for a specific work item

get_workitem_file read-only

Get file information for a specific work item attachment

create_workitem_attachment writes

Upload a local file as an attachment to a work item

list_workitem_activities read-only

List activity history for a work item (field changes, status transitions, association changes)

list_current_user_effort_records read-only

Get the user's actual effort records; the interval between end and start time cannot exceed 6 months

list_effort_records read-only

Get actual effort records

create_effort_record writes

list_estimated_efforts read-only

Get estimated effort records

create_estimated_effort writes

update_effort_record writes

Update registered actual effort

update_estimated_effort writes

Update registered estimated effort

get_pipeline read-only

Get pipeline details

list_pipelines read-only

Get pipeline list

smart_list_pipelines read-only

Smart pipeline search with natural language time references

generate_pipeline_yaml read-only

Generate pipeline YAML configuration

create_pipeline_from_description writes

Create a pipeline from natural language description

update_pipeline writes

Update pipeline YAML content

create_pipeline_run writes

Create a pipeline run instance

get_latest_pipeline_run read-only

Get the latest pipeline run instance

get_pipeline_run read-only

Get pipeline run details

list_pipeline_runs read-only

Get pipeline run list

list_pipeline_jobs_by_category read-only

Get pipeline execution tasks by category

list_pipeline_job_historys read-only

Get the execution history of a pipeline task

execute_pipeline_job_run writes

Manually run a pipeline task

get_pipeline_job_run_log read-only

Get the execution logs of a pipeline job

list_service_connections read-only

List service connections in organization

create_resource_member writes

Create a resource member

delete_resource_member destructive

Delete a resource member

list_resource_members read-only

Get a list of resource members

update_resource_member writes

Update a resource member

update_resource_owner writes

Transfer resource owner

create_tag writes

Create a tag

create_tag_group writes

Create a tag group

list_tag_groups read-only

Get a list of tag groups

delete_tag_group destructive

Delete a tag group

update_tag_group writes

Update a tag group

get_tag_group read-only

Get a tag group

delete_tag destructive

Delete a tag

update_tag writes

Update a tag

stop_vm_deploy_order writes

Stop VM deploy order

skip_vm_deploy_machine writes

Skip VM deploy machine

retry_vm_deploy_machine writes

Retry VM deploy machine

resume_vm_deploy_order writes

Resume VM deploy order

Use cases

AI-assisted code review: automatically analyze merge requests, provide comments and improvement suggestions.

Automated task management: create, update work items based on requirements, and track effort.

Pipeline management: create and run pipelines via natural language descriptions.

Code repository management: create branches, file operations, merge request management.

Test management: create test cases, update test results, maintain test plans.

Supported clients

Claude Desktop

Cursor

Listed from the project's documentation, not tested by this site.

Overview

The Alibaba Cloud DevOps MCP Server provides AI assistants with the ability to interact with the Yunxiao platform, enabling them to read work item contents in projects, automatically write code after understanding requirements, and submit code merge requests. Enterprise development teams can use it to assist with code reviews, optimize task management, reduce repetitive operations, and thus focus on more important innovation and product delivery.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →

NPM Sentinel MCP Server 79 · B

AI-powered NPM package analysis MCP server

★ 18 · Tools 19 Compare with this →

MartinLoop MCP Server 79 · B

An independent control layer that gives AI coding agents budgets, brakes, and receipts

★ 410 Compare with this →

GitHub MCP Server 76 · B

GitHub's official MCP server for managing repos, issues, PRs, and workflows via natural language

★ 33.5k · Tools 15 Compare with this →

Compare with other MCP servers

Source revision ab01844e3d6f Data synced 2026-10-12 Read the FMRS scoring method