Name: AlibabaCloud DevOps MCP
Author: aliyun

LABSCO SUMMARY

One connection answers questions that currently need three consoles.

A change request can carry its reviewers and its work items from the moment it is created, then run through a pipeline whose per-step logs you read back call by call. The size is the thing to plan for: at 197 tools, ask for the toolsets that match the job rather than loading everything, and take the hosted endpoint if you would rather not run the process at all.

What it is

Alibaba Cloud's Yunxiao platform as 197 tools, grouped by what they act on: Codeup repositories and change requests, projects with work items and sprints, pipelines and their jobs, VM deploy orders, application delivery, artifact repositories and test management.

What you get

Codeup repository work: branches created and deleted, files written on a branch with a commit message, and two refs compared
Change requests raised with reviewers and work items attached, commented on globally or inline against a file and line, and followed through their patch sets
Work items created, searched and updated, carrying comments, attachments, activity history and related test cases
Effort logged through its own calls, because Yunxiao holds actual and estimated hours as controlled fields that the work-item update cannot set
Pipelines found by natural-language time reference, run against a branch or tag, and driven job by job: retry, skip, stop, or pass and refuse a manual checkpoint
Deployment followed down to one machine — VM deploy orders stopped, resumed and retried per machine, with that machine's deploy log
Test repositories, cases and plans, with per-plan progress counted as passed, failed, postponed and pending
Applications, orchestrations and variable groups for delivery, plus package repositories and their artifacts

Requirements

A Yunxiao organization and a personal access token, granted read and write across the API groups you intend to use. Yunxiao runs it hosted at https://openapi-rdc.aliyuncs.com/ai/mcp — Bearer token, nothing installed — or on a region-edition domain of the form https://<YOUR_ORG>.devops.aliyuncs.com/ai/mcp. Self-hosting instead needs Node 18 or newer and YUNXIAO_ACCESS_TOKEN, plus YUNXIAO_API_BASE_URL when you are on a region edition. Because the catalogue is large, both the hosted endpoint and the container build let you ask for only the toolsets you need.

Setup effort

One command plus a key — npx -y alibabacloud-devops-mcp-server, then supply credentials

See installation options

USE CASES

Getting a merge request reviewed without leaving the editor

Open the change request with its reviewers and work items already attached, then read and answer review comments — global ones and the inline ones tied to a file and line — from the same conversation the code was written in.

create_change_requestget_change_requestcreate_change_request_commentlist_change_request_comments

Working out why last night's pipeline failed

Find the run by asking for pipelines from 'today' or 'this week', walk into the failing job's steps, pull the log for the step that broke, and retry that job once the cause is clear.

smart_list_pipelinesget_latest_pipeline_runget_pipeline_job_stepsget_pipeline_job_step_logretry_pipeline_job_run

Planning a sprint from the work items that are actually open

Search the backlog by assignee, status and priority, see which sprints exist, open the next one with dates and a capacity, and move the chosen items into it.

search_workitemslist_sprintscreate_sprintupdate_work_item

Logging the hours a task actually took

Effort has its own calls because Yunxiao locks those fields: actual and estimated hours are controlled, so the ordinary work-item update will not set them however the request is shaped.

create_effort_recordcreate_estimated_effortlist_current_user_effort_records

Getting a stalled VM deployment moving again

Read the deploy order, pull the log for the machine that is stuck, then retry that machine or skip it so the rest of the rollout continues.

get_vm_deploy_orderget_vm_deploy_machine_logretry_vm_deploy_machineskip_vm_deploy_machine

Reporting how far a test plan has actually got

Progress comes back already counted as passed, failed, postponed and pending, so the status update is a read rather than a tally, and a result can be corrected in the same pass.

list_test_plansget_test_result_listget_test_plan_progressupdate_test_result

197 tools

Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.

`add_host_list_to_deploy_group`

[application delivery] Add host list to deploy group

INSTALLATION

Connect your client

Server setup

npx

TERMINAL

npx -y alibabacloud-devops-mcp-server

Hosted endpoint

ENDPOINT URL

https://openapi-rdc.aliyuncs.com/ai/mcp

Add this hosted endpoint to your MCP client using the credentials listed under Requirements.

MCP Clients

Any MCP client — Remote

ANY MCP CLIENT — REMOTE · JSON

{
  "mcpServers": {
    "yunxiao": {
      "url": "https://openapi-rdc.aliyuncs.com/ai/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      }
    }
  }
}

Any MCP client — Remote region

ANY MCP CLIENT — REMOTE REGION · JSON

{
  "mcpServers": {
    "yunxiao": {
      "url": "https://<YOUR_ORG>.devops.aliyuncs.com/ai/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      }
    }
  }
}

Any MCP client — npx

ANY MCP CLIENT — NPX · JSON

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

Any MCP client — npx region

ANY MCP CLIENT — NPX REGION · JSON

{
  "mcpServers": {
    "yunxiao": {
      "command": "npx",
      "args": [
        "-y",
        "alibabacloud-devops-mcp-server"
      ],
      "env": {
        "YUNXIAO_ACCESS_TOKEN": "<YOUR_TOKEN>",
        "YUNXIAO_API_BASE_URL": "https://<YOUR_ORG>.devops.aliyuncs.com"
      }
    }
  }
}

Any MCP client — Docker stdio

ANY MCP CLIENT — DOCKER STDIO · JSON

{
  "mcpServers": {
    "yunxiao": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "YUNXIAO_ACCESS_TOKEN",
        "build-steps-public-registry.cn-beijing.cr.aliyuncs.com/build-steps/alibabacloud-devops-mcp-server:latest"
      ],
      "env": {
        "YUNXIAO_ACCESS_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}

AlibabaCloud DevOps MCP

One connection answers questions that currently need three consoles.

What it is

What you get

Requirements

Setup effort

Getting a merge request reviewed without leaving the editor

Working out why last night's pipeline failed

Planning a sprint from the work items that are actually open

Logging the hours a task actually took

Getting a stalled VM deployment moving again

Reporting how far a test plan has actually got

197 tools

add_host_list_to_deploy_group

add_host_list_to_host_group

cancel_app_release_stage_execution

cancel_appstack_change_request

close_appstack_change_request

compare

create_app_orchestration

create_app_tag

create_application

create_appstack_change_request

create_branch

create_change_order

create_change_request

create_change_request_comment

create_commit_comment

create_effort_record

create_estimated_effort

create_file

create_flow_variable_group

create_global_var

create_pipeline_from_description

create_pipeline_run

create_repository

create_resource_member

create_sprint

create_test_plan_testcase_comment

create_testcase

create_testcase_comment

create_testcase_directory

create_variable_group

create_version

create_work_item

create_work_item_comment

create_workitem_attachment

create_workitem_testcase_relation

delete_app_orchestration

delete_branch

delete_file

delete_flow_variable_group

delete_resource_member

delete_testcase

delete_variable_group

delete_version

delete_work_item

delete_workitem_testcase_relation

execute_app_release_stage

execute_job_action

execute_pipeline_job_action

execute_pipeline_job_run

find_task_operation_log

generate_pipeline_yaml

get_app_orchestration

get_app_release_stage_job_log

get_app_release_stage_pipeline_run

get_app_release_workflow_stage

get_app_variable_groups

get_app_variable_groups_revision

get_application

get_appstack_change_request_audit_items

get_artifact

get_branch

get_change_order

get_change_request

get_commit

get_current_organization_info

get_current_user

get_env_variable_groups

`add_host_list_to_deploy_group`

`add_host_list_to_host_group`

`cancel_app_release_stage_execution`

`cancel_appstack_change_request`

`close_appstack_change_request`

`compare`

`create_app_orchestration`

`create_app_tag`

`create_application`

`create_appstack_change_request`

`create_branch`

`create_change_order`

`create_change_request`

`create_change_request_comment`

`create_commit_comment`

`create_effort_record`

`create_estimated_effort`

`create_file`

`create_flow_variable_group`

`create_global_var`

`create_pipeline_from_description`

`create_pipeline_run`

`create_repository`

`create_resource_member`

`create_sprint`

`create_test_plan_testcase_comment`

`create_testcase`

`create_testcase_comment`

`create_testcase_directory`

`create_variable_group`

`create_version`

`create_work_item`

`create_work_item_comment`

`create_workitem_attachment`

`create_workitem_testcase_relation`

`delete_app_orchestration`

`delete_branch`

`delete_file`

`delete_flow_variable_group`

`delete_resource_member`

`delete_testcase`

`delete_variable_group`

`delete_version`

`delete_work_item`

`delete_workitem_testcase_relation`

`execute_app_release_stage`

`execute_job_action`

`execute_pipeline_job_action`

`execute_pipeline_job_run`

`find_task_operation_log`

`generate_pipeline_yaml`

`get_app_orchestration`

`get_app_release_stage_job_log`

`get_app_release_stage_pipeline_run`

`get_app_release_workflow_stage`

`get_app_variable_groups`

`get_app_variable_groups_revision`

`get_application`

`get_appstack_change_request_audit_items`

`get_artifact`

`get_branch`

`get_change_order`

`get_change_request`

`get_commit`

`get_current_organization_info`

`get_current_user`

`get_env_variable_groups`

`get_file_blobs`

`get_flow_variable_group`

`get_global_var`

`get_latest_orchestration`

`get_latest_pipeline_run`

`get_machine_deploy_log`

`get_organization_department_ancestors`

`get_organization_department_info`

`get_organization_member_info`

`get_organization_member_info_by_user_id`

`get_organization_role`

`get_pipeline`

`get_pipeline_job_run_log`