ShadowGate MCP

ShadowGate MCP

Defensive MCP security gateway for AI agents. ShadowGate gates MCP tool calls, responses, manifests, trust identity, drift detection, and security reports before agents trust or execute external MCP …

Quick Start

# Connect this server (installs CLI if needed)
npx -y smithery mcp add jossephibrahim/shadowgate-mcp

# Browse available tools
npx -y smithery tool list jossephibrahim/shadowgate-mcp

# Get full schema for a tool
npx -y smithery tool get jossephibrahim/shadowgate-mcp health_check

# Call a tool
npx -y smithery tool call jossephibrahim/shadowgate-mcp health_check '{}'

Direct MCP Connection

Endpoint: https://shadowgate-mcp--jossephibrahim.run.tools

Required config:

• client_key (query) — Client key required for protected ShadowGate scan/gateway tools. Do not share publicly

Optional config:

• admin_key (query) — Optional admin key for private admin tools. Leave empty for normal use.

Tools (26)

• health_check — Return ShadowGate MCP health, version, and active policy.
• scan_text — Scan arbitrary text for leaked secrets, prompt injection, risky commands, and sensitive file paths.
• analyze_text
• redact_secrets — Return the text with detected secrets and sensitive path snippets redacted.
• get_risk_score — Return a 0-100 risk score for a text payload.
• decide_policy — Return the policy decision for a payload: allow, redact, or block.
• inspect_mcp_response — Scan a response returned by another MCP server before the agent trusts it.
• inspect_mcp_tool_call — Scan an outgoing MCP tool call before execution.
• gate_mcp_tool_call — Gateway decision for an outgoing MCP tool call.
• gate_mcp_response — Gateway decision for an MCP response.
• evaluate_mcp_transaction
• inspect_tool_schema — Scan an MCP tool schema/description before allowing agents to use it.
• review_mcp_manifest
• scan_batch
• simulate_policy_modes — Show how the same text would be handled under monitor, balanced, and strict modes.
• get_policy — Return the active ShadowGate policy configuration.
• set_policy_mode — Change ShadowGate policy mode: monitor, balanced, or strict.
• get_recent_audit_events — Return recent ShadowGate audit events. Raw scanned text is never stored.
• get_audit_summary — Return a summary of ShadowGate audit decisions, actions, categories, and severities.
• create_security_report — Create a compact security report from recent audit events.
• get_server_registry — Return the ShadowGate MCP server trust registry.
• get_mcp_server_trust — Return trust status for a specific MCP server.
• set_mcp_server_trust
• approve_mcp_manifest_identity — Admin tool to approve and persist a manifest trust identity baseline.
• get_data_paths — Return ShadowGate data directory paths for policy, registry, and audit logs.
• get_security_config — Return ShadowGate admin-auth security configuration without exposing the raw key.

# Get full input/output schema for a tool
npx -y smithery tool get jossephibrahim/shadowgate-mcp