MCP tools in GitHub Copilot

developer2026-11-10

On this page

GitHub Copilot calls MCP tools after you add an MCP server to its editor configuration. In VS Code, the server definition lives in a JSON file in your workspace. Copilot lists the server's tools in agent mode and calls them during a chat session.

What MCP adds to GitHub Copilot

MCP is an open protocol that lets an AI client call external tools through one standard interface. GitHub Copilot acts as the client. An MCP server supplies the tools.

Copilot reads the tool list, chooses a tool, and sends arguments. The server runs the action and returns a result.

Without MCP, Copilot works from the files in your workspace and its built-in knowledge. With MCP, the agent can query a database, read a ticket, or call an API your team already uses. The model decides when to call each tool. You decide which servers stay connected.

Set up an MCP server in VS Code

Add a server definition to a workspace file. In VS Code, that file is .vscode/mcp.json. Each entry names the server and gives the command or URL that starts it.

Open the file and add the server block. Save the file. Then open Copilot Chat in agent mode and check the tools list. The server appears there when the connection succeeds.

  • Use a remote URL when a provider hosts the server.
  • Use a local command when you run the server on your own machine.

GitHub documents the current setup steps at https://docs.github.com/en/copilot. Read that page when the editor menus change.

Use one gateway for many providers

Each provider usually has its own key, base URL, and rules. A gateway places those providers behind one MCP endpoint. Copilot then sees one server and a tool list drawn from many providers.

The APIbase gateway works this way. The catalog covers hundreds of providers across many categories. A team can find tools for search, data, and messaging without adding a server for each one. Each tool still needs review before the team relies on it.

Pick tools that fit agent work

Agents perform best with a small set of clear tools. Each tool needs a plain name, a short description, and typed inputs. Copilot reads these fields to decide when to call a tool.

Start with read-only tools. Read-only calls carry less risk than writes. Add write tools after your team reviews the outputs they return.

Keep credentials out of the config file

Do not paste API keys into .vscode/mcp.json. If you commit that file to git, anyone with repository access sees the key. Store secrets in an environment variable or the editor secret store. Reference the variable from the config.

Rotate any key that appears in a commit. Check the git history before you push.

Test a tool before the team relies on it

Run each tool once with a harmless input. Compare the output with a direct call to the provider API. Note the latency and the error format.

Write a short test prompt for each tool. Ask Copilot to call the tool and report the result. A failing prompt often points to a weak description or an unclear input schema.

Plan the cost of MCP tool calls

Gateway fees and Copilot subscription fees are separate costs. APIbase plan prices fall between 0 and 29.99 USD. Check the pricing page for current terms before you budget.

GitHub bills Copilot under its own plans. No public comparison of gateway fees was found, so this guide makes no claim about rival pricing.

Check the setup when a tool does not appear

Start with the server log. Most failed connections show an error there. Confirm that the local command runs in a terminal. Confirm that the remote URL responds to a plain request.

  • Restart the editor after you edit the config file.
  • Confirm that agent mode is on in Copilot Chat.

Live pricing — developer

Price per call, developer tools Horizontal bar chart of the price per call and the cache-hit price, in US dollars, for 12 developer tools. The same values are in the table below the chart. Price per call Cache-hit price $0.0 $0.0005 $0.001 Usage Time SeriesUsage Time Series $0.0 $0.0 Per-Tool Usage BreakdownPer-Tool Usage Breakdown $0.0 $0.0 Usage SummaryUsage Summary $0.0 $0.0 Discover ToolsDiscover Tools $0.0 $0.0 Batch Tool CallsBatch Tool Calls $0.0 $0.0 Tool Quality MetricsTool Quality Metrics $0.0 $0.0 Tool Quality RankingsTool Quality Rankings $0.0 $0.0 List Programming LanguagesList Programming Languages $0.001 $0.0001 Check CVE ID Reservation Stat…Check CVE ID Reservation Status $0.001 $0.0001 Security Advisories (deps.dev)Security Advisories (deps.dev) $0.001 $0.0001 Dependency Tree (deps.dev)Dependency Tree (deps.dev) $0.001 $0.0001 Package Info (deps.dev)Package Info (deps.dev) $0.001 $0.0001 US dollars per call
Price per call for developer tools, cheapest first.

Data source: live tool catalog snapshot, fetched 2026-11-10.

Chart data as a table
ToolPrice per callCache-hit price
Usage Time Series$0.0$0.0
Per-Tool Usage Breakdown$0.0$0.0
Usage Summary$0.0$0.0
Discover Tools$0.0$0.0
Batch Tool Calls$0.0$0.0
Tool Quality Metrics$0.0$0.0
Tool Quality Rankings$0.0$0.0
List Programming Languages$0.001$0.0001
Check CVE ID Reservation Status$0.001$0.0001
Security Advisories (deps.dev)$0.001$0.0001
Dependency Tree (deps.dev)$0.001$0.0001
Package Info (deps.dev)$0.001$0.0001
ToolProviderPrice/callCache-hit
Usage Time Seriesaccount$0$0
Per-Tool Usage Breakdownaccount$0$0
Usage Summaryaccount$0$0
Discover Toolsapibase$0$0
Batch Tool Callsplatform$0$0
Tool Quality Metricsplatform$0$0
Tool Quality Rankingsplatform$0$0
List Programming Languagesjudge0$0.001$0.0001
Check CVE ID Reservation Statuscve-mitre$0.001$0.0001
Security Advisories (deps.dev)depsdev$0.001$0.0001
Dependency Tree (deps.dev)depsdev$0.001$0.0001
Package Info (deps.dev)depsdev$0.001$0.0001

Connect via MCP

How a tool call worksAn AI agent sends a tool call to the APIbase gateway. The gateway checks payment, calls the provider API, and returns the JSON result to the agent.Your agentAPIbase gatewayProvider API1. tool call2. paid call3. JSON result flows back to the agent
Diagram: the path of one tool call.

Data source: gateway design, no measured values.

$ curl -X POST https://apibase.pro/api/v1/tools/account.timeseries/call \
  -H "Content-Type: application/json" -d '{"params": {}}'

FAQ

Does GitHub Copilot support MCP?

Yes. Copilot agent mode can call tools from MCP servers that you configure in your editor. Check the GitHub Copilot documentation at https://docs.github.com/en/copilot for the current support details.

Where do I put the MCP server config for Copilot in VS Code?

Place the server definition in .vscode/mcp.json in your workspace. Save the file, reload the window, and check the tools list in Copilot Chat.

Is it safe to share an MCP config with the team?

Share server names and URLs, but never share secret values. Keep keys in environment variables and out of the committed file.

How much does an MCP gateway cost?

APIbase plans range from 0 to 29.99 USD. Check the pricing page for current terms.

  1. STEP 1/2 Connect your MCP client
  2. STEP 2/2 Browse the full catalog

Written with AI assistance from APIbase's published documentation and public data; checked automatically against them. Corrections: https://apibase.pro/

Published: 2026-11-10Source: live apibase.pro catalog snapshot