Technical Architecture
MCP Security & Governance Deep technical reference for governing Model Context Protocol servers, stdio streams, and tools/call invocations inline.
What is MCP Security?
The Model Context Protocol (MCP) allows AI clients (Cursor, Claude Desktop, Windsurf, Zed) to dynamically connect to tools and data sources. Gaussian establishes an inline mediation layer between AI clients and MCP servers to enforce zero-latency policy controls before side effects occur.
Canonical Definition · Model Context Protocol (MCP) Security Model Context Protocol (MCP) Security in Gaussian is an inline proxy-free firewall layer for macOS that inspects stdio streams and SSE/WebSocket JSON-RPC 2.0 tools/call frames. It evaluates allowed vs. blocked tool methods, redacts secret credentials, and enforces zero-trust policy decisions.
Discovery & Inventory Engine Gaussian continuously scans developer workstations for declared and undeclared MCP server configurations across client configuration paths:
~/Library/Application Support/Cursor/User/globalStorage/mcp.json~/Library/Application Support/Claude/claude_desktop_config.json~/.codeium/windsurf/mcp_config.json~/.config/zed/settings.jsonArchitecture Flow
Inline Model Context Protocol (MCP) Mediation Layer Sub-1ms MediationPhase 01 JSON-RPC 2.0
AI Client tools/call Cursor, Claude, Windsurf dispatches JSON-RPC payload
Phase 02 OS Context
Identity & PID Attribution Binds request to client binary, parent PID, & signature
Phase 03 DLP Filter
Inline Secret Redaction Scans & strips AWS keys, SSH credentials, & PII
Phase 04 ALLOW / BLOCK
Policy Engine Verdict Evaluates allowed vs. blocked tool methods
Intercepts stdio streams, SSE, and WebSocket JSON-RPC 2.0 payloads inlineZero Custom Proxy Config
Interactive MCP Server & Tool Governance Catalog Explore representative MCP server integrations, risk tiers, allowed vs. blocked tools, and inline verdict enforcement below:
Transport: ALL stdio SSE HTTP
GitHub API integration server for code search, issue management, and PR reviews.
Mediated Payload Frame JSON-RPC 2.0
{"method": "tools/call", "params": {"name": "read_file_contents", "arguments": {"path": "src/config.py"}}}✓ Allowed Tools (3) search_repositories read_file_contents create_issue
⚠ Blocked / Restricted Tools (3) delete_repository force_push modify_branch_protection
Secret Redaction: ACTIVE Verdict: INLINE MEDIATED
postgres-mcp stdio
CRITICAL PostgreSQL database connector allowing AI clients to query database schema and tables.
Mediated Payload Frame JSON-RPC 2.0
{"method": "tools/call", "params": {"name": "DROP_DATABASE", "arguments": {"dbname": "production"}}}✓ Allowed Tools (2) query_readonly_table describe_schema
⚠ Blocked / Restricted Tools (4) DROP_DATABASE TRUNCATE_TABLE ALTER_ROLE GRANT_ALL
Secret Redaction: ACTIVE Verdict: INLINE MEDIATED
filesystem-mcp stdio
CRITICAL Local disk tool server allowing AI agents to read files in workspace directory.
Mediated Payload Frame JSON-RPC 2.0
{"method": "tools/call", "params": {"name": "read_file", "arguments": {"path": "~/.ssh/id_rsa"}}}✓ Allowed Tools (2) read_file list_directory
⚠ Blocked / Restricted Tools (3) write_file_root read_ssh_keys read_env_secrets
Secret Redaction: ACTIVE Verdict: INLINE MEDIATED
Slack notification bridge for sending build status updates and notifications.
Mediated Payload Frame JSON-RPC 2.0
{"method": "tools/call", "params": {"name": "post_channel_message", "arguments": {"channel": "#deployments"}}}✓ Allowed Tools (2) post_channel_message read_public_channels
⚠ Blocked / Restricted Tools (2) read_direct_messages invite_user
Secret Redaction: ACTIVE Verdict: INLINE MEDIATED
terminal-exec-mcp stdio
CRITICAL Local shell executor running developer build scripts and unit tests.
Mediated Payload Frame JSON-RPC 2.0
{"method": "tools/call", "params": {"name": "sudo", "arguments": {"command": "rm -rf /"}}}✓ Allowed Tools (2) run_approved_linter npm_test
⚠ Blocked / Restricted Tools (3) sudo curl_pipe_bash chmod_777
Secret Redaction: ACTIVE Verdict: INLINE MEDIATED
Brave Search API integration for real-time web search and documentation retrieval.
Mediated Payload Frame JSON-RPC 2.0
{"method": "tools/call", "params": {"name": "brave_web_search", "arguments": {"query": "Next.js docs"}}}✓ Allowed Tools (2) brave_web_search brave_local_search
⚠ Blocked / Restricted Tools (1) scrape_internal_intranet
Secret Redaction: ACTIVE Verdict: INLINE MEDIATED
MCP Transport Layer Governance Matrix Protocol Inspection stdio (Local IPC) AI clients spawn local binary or Node/Python process via stdin/stdout pipe streams.
High OS Privileges — direct access to filesystem, environment variables, local credentials, and shell execution.
Gaussian Automated Defense Process interception via Endpoint Security (ES) + stdio payload sanitization inline.
SSE (Server-Sent Events) Long-lived HTTP event stream connection to remote or local HTTP endpoints.
Data exfiltration path — remote tool execution and payload streaming out of enterprise network boundary.
Gaussian Automated Defense Network Extension (NE) flow filter + URL domain allowlisting + payload inspection.
Streamable HTTP / WebSockets Bidirectional JSON-RPC 2.0 message transport over HTTP POST / WebSocket connections.
Bypasses standard CASB & web proxies if client encrypts or tunnels request payloads.
Gaussian Automated Defense Content Filter mediation + TLS payload inspection before request payload dispatch.
Inline tools/call Mediation & Secret Redaction When an AI client invokes an MCP tool (e.g. tools/call with name exec_sql or read_credentials), Gaussian intercepts the JSON-RPC request frame:
Identity Attribution : Attributes the invocation to the specific initiating AI client process (PID, signature, bundle ID).Argument Sanitization : Scans arguments for AWS keys, SSH credentials, bearer tokens, or PII and redacts them in transit.Policy Evaluation : Evaluates allowed vs. blocked tool names against active enterprise posture.Action Firewall Decision : Emits ALLOW, BLOCK, or suspends for HUMAN_IN_THE_LOOP approval.Canonical Policy Schema Configuration mcp_governance_policy.json Gaussian PolicyEngine v2
{
"version": "v2",
"mcp_governance": {
"discovery_mode": "continuous",
"allow_undeclared_servers": false,
"servers": [
{
"name": "github-mcp",
"transport": "stdio",
"command_path": "/usr/local/bin/github-mcp-server",
"allowed_tools": ["search_repositories", "read_file_contents", "create_issue"],
"blocked_tools": ["delete_repository", "force_push", "modify_branch_protection"],
"secret_redaction": true
},
{
"name": "postgres-mcp",
"transport": "stdio",
"risk_level": "critical",
"action": "enforce_prompt_approval"
}
]
}
}Frequently Asked Questions (FAQ) What is Model Context Protocol (MCP) Security? MCP Security is the real-time mediation and policy enforcement layer that governs JSON-RPC 2.0 tools/call requests between AI clients (Cursor, Claude, Windsurf) and local or remote MCP servers.
How does Gaussian redact credentials in MCP tool calls? Gaussian inspects request argument frames inline, scanning for AWS secret keys, SSH private keys, bearer tokens, and PII, redacting sensitive parameters before sending the frame to the MCP server.