Skip to main content

MCP Connectors & Tools

TL;DR
  • Connect any MCP-compatible service — Datadog, Splunk, GitHub, New Relic, and more — in minutes
  • Select which connector tools your agent can call directly — no custom agent needed for simple tool use
  • Your agent auto-discovers tools from connected servers and monitors connection health
  • Tool capacity bar shows your budget (80 tools per agent) with color-coded warnings
  • Two transport types: Streamable-HTTP for remote services, stdio for local processes

The problem: your agent can't see half your stack

Your agent has deep built-in access to Azure services — Application Insights, Log Analytics, Azure Monitor, Resource Graph. But your operational toolchain doesn't stop at Azure. Metrics live in Datadog. Security logs live in Splunk. Source code lives in GitHub.

During an incident, that gap forces you to manually bridge systems — querying Datadog in one tab, correlating with Azure Monitor in another, copy-pasting findings between tools. Your agent is powerful, but it only sees the Azure side.

How MCP connectors work

MCP architecture diagram showing two transport types (Streamable-HTTP for remote services, Stdio for local processes), pre-configured partner connectors, authentication methods, and tool assignment to custom agents

The Model Context Protocol (MCP) is an open standard for connecting AI agents to external tools. An MCP server wraps any service — a database, an API, a monitoring platform — and exposes its capabilities as tools your agent can discover and invoke.

When you connect an MCP server to your agent:

  1. Auto-discovery — Your agent calls the server's tool listing and immediately knows what tools are available
  2. Namespaced registration — Each tool is registered with a prefixed name (e.g., my-datadog_list_metrics) so tools from different servers never conflict
  3. Health monitoring — Your agent pings each server every 60 seconds, auto-recovers from transient failures, and shows real-time connection status
  4. Dynamic updates — When a server adds new tools, your agent detects them automatically within 5 minutes

Two transport types

MCP supports two ways to connect your agent to external tools:

TransportHow it worksBest forExample
Streamable-HTTPYour agent connects to a remote URL endpointCloud-hosted services, SaaS platformsDatadog, GitHub, New Relic
StdioYour agent runs a local process that communicates via stdin/stdoutCustom servers, npm packages, sidecar processesnpx-based servers, Python MCP servers

Streamable-HTTP (remote services)

Connect to any MCP server accessible via URL. Provide the endpoint and authentication credentials — either a Bearer token or custom headers.

Stdio (local processes)

Run an MCP server as a process inside your agent's cloud environment. Provide the command, arguments, and optional environment variables. Your agent manages the process lifecycle — starting it on connection, monitoring health, and restarting on failure.

Supported runtimes for stdio

Stdio MCP servers run inside the agent's container. Available runtimes:

CommandRuntime
npx, nodeNode.js 20
python, python3Python 3.12
dotnet.NET 9

Docker containers are not supported as stdio commands. Only pre-installed runtimes are available.

Pre-configured partner connectors

Your agent includes pre-configured connectors for popular platforms. These have pre-filled URLs, default auth settings, and branded icons — select the card, enter your credentials, and connect.

PartnerTransportWhat you provideWhat your agent gets
GitHubStreamable-HTTPPersonal Access TokenRepository search, PR listing, issue tracking, code search
DatadogStreamable-HTTPAPI Key + Application KeyMetrics queries, APM traces, alert management
New RelicStreamable-HTTPAPI KeyPerformance monitoring, analytics
SplunkStreamable-HTTPBearer tokenLog search, event queries
DynatraceStreamable-HTTPBearer tokenApplication performance monitoring
ElasticsearchStreamable-HTTPAuthorization headerSearch, analytics, log management
Hawkeye (NeuBird)Streamable-HTTPBearer tokenAI-powered observability

For partner connectors, the authentication method is pre-configured and locked — you just enter the credential values.

Finding more MCP servers

Browse Azure MCP Center for verified MCP servers for Azure services.

How tools reach your agents

MCP tools can reach your agent in two ways:

  1. Agent tools — Select tools during connector setup or editing. These tools are available directly in the main conversation — no custom agent needed.
  2. Custom agent tools — Assign tools to specific custom agents for focused specialists with domain expertise.

Agent tools

When you create or edit an MCP connector, a tool selection step lets you choose which tools are visible to your agent. Selected tools are injected into your agent's tool list and kept in sync — when you add, remove, or update connectors, your agent's tools refresh automatically.

During connector creation: After your connector connects successfully, a Select tools step appears. All discovered tools are pre-selected up to remaining capacity. Click Done to save or Skip to add tools later.

For existing connectors: Edit any MCP connector to find the MCP Tools section at the bottom of the dialog. Check or uncheck tools to control which ones your agent can call.

No custom agent required

With your agent tool visibility, you can connect an MCP server, select tools, and start asking questions — all in three steps.

Custom agent tools

For focused specialists, assign MCP tools to specific custom agents via the portal or YAML.

Portal: In Builder > Agent Canvas, edit an agent → Advanced settings > ToolsChoose tools.

YAML wildcards:

Custom agent with wildcard MCP tools
mcp_tools:
- datadog-mcp/* # All tools from this connection
- github_search_code # One specific tool

The {connection-id}/* pattern adds every tool from that server, including tools added later. The forward slash is required.

ApproachWhen to use
Individual toolsPrecise control over tool access
Wildcard (connection-id/*)All tools from a server, including future ones
MixedAll from one server, specific picks from another
Your agent and custom agent tools are independent

The same tool can be visible to both your agent and a custom agent. There is no conflict — the tool is available in both contexts.

For a full walkthrough, see Set Up MCP Tools →.

Tool selection and capacity

Each agent — whether your agent or a custom agent — can use up to 80 tools (native and MCP combined). The tool selection UI helps you manage this budget across connectors.

Capacity indicator

The tool picker shows a progress bar with your current tool count:

Tool countBar colorMeaning
0–56 (≤ 70%)BluePlenty of room
57–72 (71–90%)YellowApproaching the limit
73–80 (> 90%)RedNear or at capacity

A hint below the bar reads: "Maximum of 80 tools allowed to avoid degrading agent performance."

Behavior at capacity

When your selected tools reach the 80-tool limit:

  • Unchecked tools become disabled — you cannot add more until you remove some
  • Already-checked tools remain enabled — you can always deselect them to free capacity
  • Select all caps at remaining capacity — if you have room for 10 more tools and click Select all, only 10 are selected

The tool count spans all connectors. If Connector A has 60 tools visible to your agent, Connector B can add up to 20 more.

Connection health monitoring

Your agent continuously monitors every MCP connection:

StatusWhat it meansAction needed
ConnectedServer healthy, tools readyNone
DisconnectedTemporary loss — auto-recovery in progressUsually none — wait for next heartbeat
FailedUnrecoverable errorCheck URL, credentials, network
InitializingConnection being establishedWait
ErrorNo running agent instanceStart the agent

Your agent pings each server every 60 seconds. Transient failures recover automatically on the next successful heartbeat. Before invoking any MCP tool, your agent validates the connection and attempts reconnection if needed.

Connections persist through failures

If an MCP server goes offline, the connector stays visible with its error status. Click See details to view the error message, tool count, and last heartbeat timestamp.

Authentication

Each MCP server has its own authentication requirements. The portal supports:

Auth methodWhen to useHow it works
Bearer tokenMost SaaS APIs (GitHub, Splunk, Dynatrace)Sends token in the Authorization header
Custom headersAPIs requiring specific headers (Datadog)Sends arbitrary key-value headers with each request
Managed identityAzure services via stdio connectorsUses your agent's managed identity for Azure AD tokens

For partner connectors, the auth method is pre-configured — you just enter the credentials.

What makes this different

Unlike custom integrations, MCP connectors use an open standard. You don't write adapter code — any MCP-compatible server works out of the box. The protocol handles discovery, invocation, and error handling.

Unlike static tool configurations, your agent discovers tools dynamically. When an MCP server adds a new tool, your agent picks it up automatically. Wildcard patterns (datadog-mcp/*) ensure new tools are available without reconfiguration.

Unlike scripts and runbooks, MCP connections are self-healing. Your agent monitors every connection with a 60-second heartbeat, auto-recovers from transient failures, and defers custom agent loading until connections establish.

Before and after

Before MCPAfter MCP
Tools during investigation4+ browser tabs openSingle conversation
Using connector toolsCreate custom agent → assign tools → invoke via /agentSelect tools on connector → ask your agent directly
Data correlationManual copy-paste between systemsAgent correlates across all connected systems
Time to gather context15–30 minutes2–5 minutes
Adding new toolsWait for built-in integration or write custom scriptsConnect any MCP server in minutes
Tool capacity managementHit 80-tool limit with cryptic server errorCapacity bar with color-coded warnings
Failure handlingScripts break when APIs changeHealth monitoring with auto-recovery

Get started

ResourceWhat you'll learn
Set up an MCP connector →Step-by-step tutorial for connecting remote and local MCP servers
Install a plugin from the marketplace →Import skills and MCP server configs from community plugin marketplaces
CapabilityWhat it adds
Plugin Marketplace →Browse and install community plugins that include MCP server configurations
External observability →Use MCP connectors to query Datadog, Splunk, and other platforms during investigations
Agent Playground →Test custom agents with MCP tools before production use
Scheduled tasks →Combine MCP tools with scheduled automation
Workflow automation →Build automated workflows using MCP tools
Connectors →How all connectors work in SRE Agent
Tools →The full tool taxonomy
Custom agents →Create specialists with focused tool sets
Was this page helpful?