Skip to content

Back to the catalog

GitHub Secret Scanning MCP Server
github.blogarticle

GitHub Secret Scanning MCP Server

GitHub announced general availability of secret scanning as a tool inside the GitHub MCP server, letting MCP-compatible AI coding agents such as GitHub Copilot CLI and VS Code scan uncommitted changes for exposed credentials before a commit or PR is opened. The GA release adds enforcement of existing push-protection customizations so detection and bypass behavior configured at the repo or org level stays consistent when triggered through an agent instead of a normal git push. Requires GitHub Secret Protection enabled on the repo; an optional Advanced Security plugin adds a tailored secret-scanning workflow in Copilot Chat.

TL;DR GitHub's MCP server now offers GA secret scanning so AI coding agents can catch leaked credentials pre-commit, honoring existing push-protection rules.
Open original ↗
Document

What it is

Secret scanning as a tool in the GitHub MCP server, generally available since this changelog (dated 2026-05-05 in the URL), after public preview since March 2026. An MCP-compatible agent (GitHub Copilot CLI, VS Code) scans uncommitted changes for exposed secrets before a commit or PR. The summary covers the "what" — what it omits are the exact commands to enable and invoke it.

What to do

Set up the GitHub MCP server in your dev environment first. Then, for the optional GitHub Advanced Security plugin that gives a more tailored experience:

In GitHub Copilot CLI:

/plugin install advanced-security@copilot-plugins

In VS Code: install the advanced-security agent plugin, then begin a Copilot Chat prompt with:

/secret-scanning

To run a scan, prompt the agent directly. The page's example:

Scan my current changes for exposed secrets and show me the files and lines I should update before I commit.

Preconditions

Repository must have GitHub Secret Protection enabled. The Advanced Security plugin is optional; the base MCP secret-scanning tool works without it.

Caveats

The page states no metrics, no coverage detail on which secret types are detected, and no behavior for the non-plugin path beyond the natural-language prompt. The one substantive GA change over preview — push-protection customizations now being honored through the agent path — is already in the catalog summary; the install/invoke commands above are the only detail it lacks.

Technologies named here

Related in DevSecOps