GitGuardian MCP Server enables your AI agent to scan projects using GitGuardian's API, featuring over 500 secret detectors to prevent credential leaks before they reach public repositories. Resolve security incidents without context switching to the GitGuardian console, leveraging rich contextual data to accelerate remediation and enable automated removal of hardcoded secrets. The server supports multiple authentication methods depending on deployment: OAuth (default for stdio transport) for interactive usage, Personal Access Tokens for non-interactive environments, and per-request HTTP/SSE authentication for server deployments. It can operate in stdio mode for local IDE integrations or expose an HTTP transport by enabling MCP_PORT, with OAuth disabled in HTTP/SSE mode. Configuration supports various GitGuardian instances (SaaS, EU, self-hosted) via environment variables such as GITGUARDIAN_URL and GITGUARDIAN_CLIENT_ID. Installation and configuration examples are provided for multiple editors (Cursor, Claude Desktop, Windsurf, Zed) and self-hosted deployments. Core MCP features include secret scanning, incident management, honeytokens, authentication management, and token management.