Free IDE extension for risk-free vibe coding — keep secrets out of AI.
Control
POST /api/mcp/tools/snapshotMonitor Model Context Protocol (MCP) servers for unauthorized tool changes, capability drift, and configuration tampering. Detect when an MCP server's tool definitions change unexpectedly.
Complete these steps in order. The dashboard link opens the exact product workspace for this service.
Sign in, select the project you want to protect, then open MCP Drift.
When you first connect an MCP server, its tool definitions are captured as a baseline.
Use POST /api/mcp/tools/snapshot from server-side code. Keep API keys out of browsers and client bundles.
Run one safe case and one expected-risk case. Confirm the result in MCP Drift, then review Guard Logs before enabling the control for production traffic.
MCP servers dynamically expose tools to AI agents. If an MCP server is compromised or its configuration changes, it can expose new dangerous tools or modify existing ones. MCP Drift detects these changes in real-time, alerting you before agents can exploit them.
A step-by-step breakdown of how MCP Drift protects your AI.
When you first connect an MCP server, its tool definitions are captured as a baseline.
Tool definitions are periodically re-fetched and compared against the baseline.
Any change — new tools, modified parameters, removed tools — is flagged as drift.
Each drift is risk-scored based on the nature and severity of the change.
Capture MCP server tool definitions on first connection
Instant notifications when tool definitions change
Complete version history of all MCP server tool definitions
Automated risk assessment of each detected change
This service is implemented at the application route below. Use the tested SDK and framework guides for copy-paste integration examples.
POST /api/mcp/tools/snapshot. API keys must stay server-side. See the REST API guide and related documentation before integrating.