Enter your email and we'll send you a secure sign-in link. No password needed.
Don't have an account yet? View pricing
The link expires in 30 minutes.
This link has expired or is invalid. Request a new one below.
A workspace connects BranchDeploy to one Azure DevOps organisation. Pro, the deployment audit log, release notes, the Teams bot and AI assistants all run from it. New workspaces start with 14 days of Pro, no card needed.
Loading…
| Viewer | Editor | Admin | Owner | |
|---|---|---|---|---|
| View deployments, projects and the audit log | ✓ | ✓ | ✓ | ✓ |
| View release notes | ✓ | ✓ | ✓ | ✓ |
| Write and publish release notes | ✓ | ✓ | ✓ | |
| View billing | ✓ | ✓ | ||
| Members, credentials, branding and feed settings | ✓ | ✓ | ||
| Buy, change or transfer the subscription | ✓ | |||
| Add or remove owners | ✓ |
Owners also need to be an Azure DevOps organisation administrator to change billing.
Your subscription, what’s included, and when it renews.
Loading your subscription…
Each credential does a different job, and none depends on the others. Set up the ones you need, in any order.
Whole workspace · set up by an admin
Loading…
Copy this token now — it will not be shown again.
Linking happens from inside Azure DevOps, so only people with access there can do it. If the organisation is already linked to another BranchDeploy workspace, that workspace's owners are emailed to approve the move. It goes ahead after 7 days unless they decline, and their audit history and billing stay with them.
Whole workspace · shared by everyone
We recommend creating it from a service account. Otherwise every Teams and AI deploy shows in Azure DevOps as the person the PAT belongs to, and stops working if they leave.
Link this workspace to Azure DevOps first: from inside your Azure DevOps organisation, then come back to save a PAT.
Create it in Azure DevOps, scoped to this organisation, with:
"Execute" is the Azure DevOps scope that allows MCP to queue a pipeline run — it doesn't run individual steps. Extension Data lets AI assistants see when a teammate is holding an environment, so they can tell you before you deploy over it. The current Teams workflow only needs Work Items and Code read access to resolve branches.
Personal · just you
AI deploys also need the workspace PAT. to finish setup.Ask a workspace admin to save one.
This key will not be shown again. Copy it now and store it safely.
Each member's personal MCP API key in this workspace. Revoking one stops it straight away; that person can generate a new one. Removing someone from the workspace stops their key too.
AI assistants connected by signing in with BranchDeploy, such as ChatGPT and Claude. Each connection belongs to the person who signed in. Disconnecting stops the app straight away; it has to ask again, and be approved, to reconnect.
Loading…
BranchDeploy is a remote MCP server at https://mcp.branch-deploy.dev/mcp, authenticated
with your MCP API key as a bearer token, or by signing in with BranchDeploy for ChatGPT and Claude. Any MCP-compatible
client works — setup for the common ones is below.
Claude (claude.ai, the Desktop app and mobile) connects as a custom connector and signs in with your BranchDeploy account, so there's no config file and no API key to paste.
https://mcp.branch-deploy.dev/mcp as the URL.Connectors added on claude.ai also appear in the Desktop and mobile apps. Claude asks before every deploy or publish, and BranchDeploy refuses a deploy that wasn't previewed first.
Using the MCP API key instead
Claude Desktop's config file only accepts local (stdio) servers, so with an API key BranchDeploy connects through the
mcp-remote bridge. It needs Node.js installed for npx.
~/Library/Application Support/Claude/claude_desktop_config.json (macOS)
%APPDATA%\Claude\claude_desktop_config.json (Windows)
branchdeploy should be listed.{
"mcpServers": {
"branchdeploy": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://mcp.branch-deploy.dev/mcp",
"--header",
"Authorization:${AUTH_HEADER}"
],
"env": {
"AUTH_HEADER": "Bearer your-mcp-api-key"
}
}
}
}
The key goes in env, not inline after --header, so the space in
Bearer … isn't split into a separate argument.
Claude Code talks to remote MCP servers directly. One command registers it.
-s user to register it for every project instead of just this one.claude mcp list./mcp to see the BranchDeploy tools.claude mcp add --transport http branchdeploy https://mcp.branch-deploy.dev/mcp \
--header "Authorization: Bearer your-mcp-api-key" If the BranchDeploy ChatGPT app is available for your account or workspace, add it inside ChatGPT and sign in — no config file and no API key to paste. The MCP API key is for the other clients.
ChatGPT signs in with your BranchDeploy account, so it doesn't use the MCP API key above — but the account still needs a Pro licence and a saved Azure DevOps PAT. If you do not see BranchDeploy in ChatGPT yet, use one of the API-key MCP clients above. Deploys ask for your confirmation before anything is queued, exactly like the other clients.
VS Code reads MCP servers from .vscode/mcp.json — note it uses the
servers key, not mcpServers.
.vscode/mcp.json in your workspace with the snippet below.{
"servers": {
"branchdeploy": {
"type": "http",
"url": "https://mcp.branch-deploy.dev/mcp",
"headers": {
"Authorization": "Bearer your-mcp-api-key"
}
}
}
} .vscode/mcp.json is usually committed to the repo. To avoid checking your key in,
use a ${input:…} placeholder so VS Code prompts for it, or add the server globally
with MCP: Add Server from the command palette.
Cursor supports remote MCP servers directly, globally or per project.
~/.cursor/mcp.json for every project, or .cursor/mcp.json inside one project.branchdeploy shows as connected.{
"mcpServers": {
"branchdeploy": {
"url": "https://mcp.branch-deploy.dev/mcp",
"headers": {
"Authorization": "Bearer your-mcp-api-key"
}
}
}
}
Any client that supports remote Streamable HTTP MCP servers works — Windsurf, Cline, Zed and the rest
all follow the same URL + bearer-header pattern. Older clients that only speak SSE can use
https://mcp.branch-deploy.dev/sse instead.
Transport: Streamable HTTP
URL: https://mcp.branch-deploy.dev/mcp
Header: Authorization: Bearer your-mcp-api-key Clients cache the tool list from the moment they connect. After changing this config, restart the client fully rather than reloading the window.
"Deploy ticket 2372""Deploy ticket 4185 to Staging" name an environment only if the project has several"Show me recent deployments"Just give the ticket number — BranchDeploy resolves its linked branch and target pipeline, applies your environment's branch rules, and asks you to confirm before queuing. It works exactly like the deploy button on the work item. Full guide: MCP documentation.
See what the Azure DevOps extension has synced, which environments are enabled, and what each deploy button will queue.
Loading…
Request, confirm, and audit BranchDeploy deployments from a Teams channel by @mentioning BranchDeploy.
The bot works in Azure DevOps through the workspace PAT saved in , for everyone who uses it in Teams.
Download the app package, then a Teams admin uploads it in Teams Admin Center → Teams apps → Manage apps → Upload a custom app. Setup guide
Download Teams app (.zip)Once installed, @mention BranchDeploy in any channel. It will list your configured environments, show confirmation cards, and record approved deployments without leaving Teams.
Turn deployments into customer-facing "What's new" copy, then publish it to a read-only public JSON feed your app can fetch. Nothing is visible to customers until you publish it.
Pick a release note from the list, or start a new draft. Nothing reaches customers until you publish it.
Write for customers, not developers. Avoid branch names, work item IDs, and implementation detail.
Deployments from the last 14 days. Adding one copies its work item title into a new item for you to rewrite, and links it as a source.
Loading…
Loading…
Applies to the hosted page only. Your logo must be hosted somewhere public over HTTPS.
Lets an AI assistant connected over MCP read the Azure DevOps ticket descriptions behind your deployments, so it can draft release notes from what the work actually was rather than from ticket titles alone.
Every deployment within your plan's audit window (90 days on Pro). Click a row for the full detail. Pipeline results (succeeded, failed) appear within a few minutes when an Azure DevOps PAT with Build (Read) access is saved on the Credentials tab; otherwise deployments show as queued.
Loading…
Loading…