Azure DevOps Advanced

ENGINEERING INTELLIGENCE

Delivery context, not just API calls.

Repositories, pull requests, boards, pipelines, tests, wiki, and security with compound engineering insights.

Azure DevOps Advanced is a Quilr-built MCP for Azure DevOps Services (dev.azure.com). It provides 47 tools across organizations, projects, Azure Repos, pull requests, Boards, Pipelines, Test Plans, Wiki, Search and Advanced Security. Each user signs in with their own Microsoft account through Microsoft Entra delegated OAuth, so Azure DevOps applies that user's existing permissions.

MCP client -> QuilrAI Gateway -> Azure DevOps MCP -> Microsoft Entra -> Azure DevOps

Azure DevOps Services only

This MCP covers cloud-hosted Azure DevOps Services. It does not cover Azure DevOps Server hosted on your own network.

Tools

Direct tools cover frequent operations with explicit schemas. Dispatcher tools group related operations behind an action selector. Intelligence tools combine several API calls into one briefing. Destructive tools always require an exact confirmation phrase.

Organization and common direct tools

Tool	Risk	Purpose
ado_list_organizations	Read	Refresh and list Azure DevOps organizations linked to the signed-in user, including the saved selection.
ado_select_organization	Read/settings	Validate and save one linked organization as the user's default.
ado_list_projects	Read	List projects with optional name/state filters and continuation pagination.
ado_list_repositories	Read	List repositories in a project, with optional filtering and hidden-repository inclusion.
ado_get_file	Read	Read a repository file from a branch, tag, or commit.
ado_list_pull_requests	Read	List pull requests across a project or within one repository using status, creator, reviewer, and target filters.
ado_get_pull_request	Read	Retrieve one pull request with optional commits and linked work items.
ado_create_pull_request	Write	Create a pull request with title, description, reviewers, draft state, and linked work items.
ado_get_work_item	Read	Retrieve one work item with selected fields and relation expansion.
ado_query_work_items	Read	Execute a bounded WIQL query and return matching work-item references.
ado_create_work_item	Write	Create a work item from an Azure DevOps field mapping and optional relations.
ado_run_pipeline	Write	Preview or queue a YAML pipeline run with variables, template parameters, resources, and skipped stages.

Compatibility dispatcher tools

Tool	Risk	Purpose and supported actions
ado_core_read	Read	Read projects, project teams, identity IDs, or connection metadata. Actions: list_projects, list_project_teams, get_identity_ids, get_connection_data.
ado_repositories_read	Read	Inspect repositories, branches, commits, directories, and files. Actions: list_repositories, get_repository, list_branches, list_my_branches, get_branch, search_commits, list_directory, get_file_content.
ado_repositories_write	Write	Create a branch. Action: create_branch.
ado_pull_requests_read	Read	List and inspect pull requests, changes, labels, threads, and comments. Actions: list, list_by_commits, get, get_changes, get_labels, list_threads, list_thread_comments.
ado_pull_requests_write	Write	Create or update pull requests, reviewers, votes, labels, threads, and comments. Actions: create, update, update_reviewers, vote, update_labels, create_thread, update_thread, reply_to_comment.
ado_search	Read	Search source code, wiki content, or work items. Actions: code, wiki, work_item.
ado_work_items_read	Read	Read work items, comments, revisions, types, backlogs, iterations, saved queries, WIQL results, and attachments. Actions: get, get_batch, list_comments, list_revisions, get_work_item_type, my_work_items, get_iteration_work_items, list_backlogs, list_backlog_work_items, get_query, get_query_results, query_wiql, get_attachment.
ado_work_items_write	Write	Create or update work items, relations, artifact/PR links, and comments. Actions: create, update, update_batch, add_children, link, unlink, add_artifact_link, link_pull_request, add_comment, update_comment.
ado_work_read	Read	Read iterations, team settings, capacity, and iteration capacity. Actions: list_iterations, list_team_iterations, get_team_settings, get_team_capacity, get_iteration_capacities.
ado_work_write	Write	Create or assign iterations and update team-member capacity. Actions: create_iterations, assign_iterations, update_capacity.
ado_pipelines_build	Read	List builds, inspect status, and list build changes. Actions: list, get_status, get_changes.
ado_pipelines_build_log	Read	List build logs or retrieve log content. Actions: list, get_content.
ado_pipelines_definition	Read	List build definitions and definition revisions. Actions: list, list_revisions.
ado_pipelines_run	Read	Get one pipeline run or list runs. Actions: get, list.
ado_pipelines_artifact	Read/download	List or download pipeline artifacts. Actions: list, download.
ado_pipelines_write	Write	Queue/preview pipelines, create pipeline definitions, or update build-stage state. Actions: run_pipeline, create_pipeline, update_build_stage.
ado_test_plans_read	Read	List test plans, suites, cases, or test results from a build. Actions: list_plans, list_suites, list_cases, show_results_from_build.
ado_test_plans_write	Write	Create plans/suites/cases, attach cases to suites, or update test steps. Actions: create_plan, create_suite, add_test_cases, create_test_case, update_test_case_steps.
ado_wiki_read	Read	List and inspect wikis, pages, and page content. Actions: list_wikis, get_wiki, list_pages, get_page, get_page_content.
ado_wiki_write	Write	Create or update a wiki page with concurrency protection. Action: upsert_page.
ado_advanced_security_read	Read	List Advanced Security alerts or retrieve one alert. Actions: list_alerts, get_alert.

Intelligence tools

Tool	Risk	Purpose
ado_project_overview	Read	Build a project briefing from repositories, active pull requests, recent builds, and open work items.
ado_pull_request_readiness	Read	Summarize draft status, reviewer votes, discussions, merge state, policies, commits, and file changes.
ado_pipeline_failure_summary	Read	Explain a failed run using stage/task outcomes, issues, logs, and build status.
ado_work_item_traceability	Read	Follow parent/child, pull request, commit, build, wiki, revision, and comment relationships for one work item.
ado_sprint_health	Read	Summarize sprint state, ownership, effort, remaining work, completion, and team capacity.
ado_security_dashboard	Read	Aggregate Advanced Security alerts across repositories by severity, state, and alert type.

Destructive tools

These tools are marked destructive, are not retried automatically, and reject the request unless the exact confirmation value is supplied. Keep them disabled in Tool visibility unless the workflow is explicitly approved.

Tool	Purpose	Exact confirmation
ado_delete_branch	Delete one Git branch reference.	DELETE BRANCH
ado_delete_repository	Delete an Azure Repos repository.	DELETE REPOSITORY
ado_delete_work_item	Soft-delete or permanently destroy a work item.	DELETE WORK ITEM or DESTROY WORK ITEM
ado_delete_build	Delete one build and its retained build data.	DELETE BUILD
ado_delete_pipeline	Delete a pipeline and its builds.	DELETE PIPELINE AND BUILDS
ado_delete_wiki_page	Delete one wiki page.	DELETE WIKI PAGE
ado_delete_test_plan	Delete one test plan.	DELETE TEST PLAN
ado_delete_iteration	Delete one project iteration.	DELETE ITERATION

Organization selection

Organization arguments are optional on organization-scoped tools:

• With one linked organization, the MCP selects and saves it automatically.
• With several organizations and no saved selection, a tool returns organization_selection_required with the choices.
• Call ado_select_organization with the exact URL-name segment, such as contoso from https://dev.azure.com/contoso. Do not pass a full URL or a display label.
• An organization passed directly to another tool is a one-call override and does not replace the saved selection.

The selection is stored per signed-in user. Access and refresh tokens are not stored with it.

Setup

The MCP uses a Microsoft Entra application with the delegated user_impersonation permission for Azure DevOps. It does not use the deprecated Azure DevOps OAuth app registration, which stopped accepting new apps in April 2025.

Who	Responsibility
Your Entra administrator	Creates or approves the Entra application, redirect URI, delegated Azure DevOps permission and client secret.
Azure DevOps MCP service	Owns the Entra OAuth configuration and exposes the MCP authorization flow.
QuilrAI administrator	Adds the server to the MCP Gateway and sets tools, access and guardrails.
Each user	Signs in with Microsoft and reaches only the Azure DevOps resources they already have.

1. Create the Microsoft Entra application

If Quilr already provides a managed Entra application for your tenant, skip this step and ask your tenant administrator to approve it.

1. In the Microsoft Entra admin center, open Entra ID > App registrations > New registration.
2. Enter a name such as QuilrAI Azure DevOps MCP.
3. Choose Accounts in this organizational directory only for a single-tenant deployment, or Accounts in any organizational directory only for an approved multi-tenant deployment.
4. Under Redirect URI, select Web and enter https://azure-devops.mcp.quilr.ai/auth/callback. Click Register.
5. From Overview, copy the Application (client) ID and Directory (tenant) ID. For an approved multi-tenant application, use organizations instead of a tenant ID.
6. Open API permissions > Add a permission > APIs my organization uses, select Azure DevOps, then under Delegated permissions enable user_impersonation.
7. Grant tenant-wide admin consent if your consent policy requires it. Otherwise users are prompted when they connect.
8. Open Certificates & secrets > Client secrets > New client secret, then copy the secret Value (not the Secret ID).

Protect the secret

Microsoft shows the client secret value only once. Store it in your approved secret manager and hand it to Quilr through the authorized onboarding channel. Never put it in an MCP client file, chat, ticket, repository or URL.

The service requests these scopes. 499b84ac-1321-427f-aa17-267ca6975798 is the Azure DevOps resource ID.

499b84ac-1321-427f-aa17-267ca6975798/user_impersonation
openid
profile
offline_access

2. Check Azure DevOps prerequisites

• The Azure DevOps organization is connected to the intended Entra tenant.
• Each user is a member or guest of every organization they need, with project permissions for the tools they will call.
• Advanced Security, Test Plans, Pipelines and Wiki are enabled and licensed where those tools are needed.
• Tenant consent and Conditional Access policies permit the Entra application.

3. Add the server to the MCP Gateway

1. Go to Settings > AI Gateway > MCP Gateway and click Add MCP server. Choose Remote server.
2. Enter Name Azure DevOps Advanced, a Slug such as azure-devops, and Transport URL https://azure-devops.mcp.quilr.ai/mcp.
3. Under How the gateway signs in, keep Auto-detect (recommended), click Probe and continue and finish the steps to Create MCP.
4. Connect OAuth once as an administrator with a Microsoft account that can reach the organization, and accept the requested permissions. The server should list 47 tools.
5. In Configure, review Tool visibility, Server access and Security guardrails before users connect.

Do not enter the Entra client secret in Cursor, Claude, ChatGPT or any other client. The upstream MCP service owns it.

4. Connect a client and verify

Point clients at the server's Quilr gateway URL, never the upstream URL. A short key such as ado keeps combined server and tool names compact:

{
  "mcpServers": {
    "ado": {
      "type": "http",
      "url": "https://mcpgateway.quilr.ai/YOUR-AZURE-DEVOPS-SLUG/mcp"
    }
  }
}

For Claude, add a custom connector with the same URL and leave its OAuth Client ID and Client Secret fields empty. Start with a read-only prompt:

Use the Azure DevOps MCP. First call ado_list_organizations. If there is one
organization, use it automatically. If there are multiple, show me the choices
and wait for my selection. Then list the first 20 projects. Do not create,
update, run, vote, comment, or delete anything.

Reconnection

Routine redeployments do not require users to sign in again while the public MCP and callback URLs, the Entra tenant and application, the OAuth signing key and the stored OAuth and organization state stay the same. Users must reconnect after the client secret expires or rotates without a service update, consent is revoked, scopes change, refresh fails permanently, OAuth state is lost, the public resource URL changes, or the Entra application is replaced.

Organization membership is refreshed automatically. A routine restart can refresh linked organizations without the user signing in again.

Troubleshooting

Error or symptom	Likely cause	Fix
AADSTS9010010 resource/scopes mismatch	The MCP resource URL was forwarded to Entra as the OAuth resource.	Confirm the MCP OAuth proxy does not forward the MCP resource and requests the Azure DevOps scope above.
invalid_client	Wrong Application ID, expired or wrong secret, wrong tenant, or the Secret ID was used instead of the Value.	Verify the Entra application, tenant and current secret value.
Redirect URI mismatch	The Entra Web redirect URI differs from https://azure-devops.mcp.quilr.ai/auth/callback.	Fix the URI in Entra, save and reconnect.
Consent or login blocked	Admin consent, user assignment, Conditional Access or cross-tenant policy blocks the app.	Ask your Entra administrator to review the enterprise application and sign-in logs.
ado_list_organizations returns nothing	The account is not linked to an Azure DevOps organization, or the organization uses another tenant.	Sign in with the right work account and confirm membership.
organization_selection_required	Several organizations and no saved default.	Call ado_select_organization with the exact URL-name segment.
401 or TF400813	Token, membership, tenant linkage or Azure DevOps authorization is invalid.	Reconnect and confirm the user is active in the organization.
403 for one tool	The user lacks that resource permission or product license.	Grant the minimum permission or use an authorized account.
Users must reconnect after a redeployment	The OAuth signing key, stored OAuth state, callback or resource URL, or Entra application changed.	Restore the stable configuration and state, or have users reconnect after verifying the new deployment.
Tools stay at Loading tools	OAuth not completed, wrong URL, or capabilities not refreshed.	Use the gateway URL, finish Connect, refresh tools and restart the client entry.

References: Entra OAuth for Azure DevOps, Azure DevOps OAuth deprecation, Microsoft Azure DevOps MCP source.

Compared with the official server

Provider-native

Choose official for local IDE work

Use Microsoft's server when developers need first-party Azure DevOps operations directly inside a supported local coding environment.

• Excellent IDE-native workflow
• First-party core domain coverage

Quilr-managed

Choose Quilr for organization-wide access

Use Quilr when teams need one hosted endpoint with delegated OAuth, gateway policy, compound delivery intelligence, and controlled destructive actions.

• Remote multi-tenant operation
• Briefs, dashboards, and confirmed deletes

Decision signalLocal developer assistance favors the official server; centrally governed engineering agents favor Quilr Advanced.

Capability	Official Azure DevOps MCP	Quilr Advanced
Core Azure DevOps domains	✅	✅
Local IDE server	✅	-
Centrally hosted streamable HTTP	-	✅
Multi-tenant Entra delegated OAuth	-	✅
Compatibility dispatchers for established operations	-	✅ 21 dispatchers / 93 operations
Direct common-operation tools	✅	✅ 12 curated tools
Compound project and sprint intelligence	-	✅
Pull-request readiness and pipeline failure briefs	-	✅
Traceability and security dashboards	-	✅
Separately confirmed destructive operations	-	✅ 8 tools

The official server is excellent for local IDE use. Quilr Advanced is designed for organization-wide access through gateway policy and auditing.