Skip to content
DeepIntShield DeepIntShield

Overview

Overview

Section titled “Overview”

DeepIntShield provides OpenAI-, Anthropic-, and Gemini-compatible routes for the documented endpoints. Point each agent at the URL shape it expects (see Configuration below and the guide for your tool), then test its exact SDK version, streaming mode, tools, and error handling. That unlocks:

  • Universal Model Access: Use any provider/model configured in DeepIntShield with any agent (e.g., use GPT-5 with Claude Code, or Claude Sonnet 4.5 with Codex CLI)
  • MCP Tools Integration: Expose only the Model Context Protocol tools bound to the agent’s authenticated Virtual Key; execution remains separately authorized
  • Built-in Observability: Monitor all agent interactions in real-time through DeepIntShield’s logging dashboard
  • Load Balancing: Automatically distribute requests across multiple providers and regions
  • Advanced Features: Governance, caching, failover, and more - all transparent to your agent

CLI Agents

Section titled “CLI Agents”

LibreChat

Open-source chat client supporting multiple providers

Open →

Open WebUI

ChatGPT-like open-source chat interface

Open →

Claude Code

AI-powered coding directly in your terminal

Open →

Codex CLI

OpenAI’s powerful code generation CLI

Open →

Gemini CLI

Google’s coding assistant with advanced reasoning

Open →

Qwen Code

Alibaba’s coding assistant

Open →

Opencode

Terminal-based coding assistant by SST

Open →

Editors

Section titled “Editors”

Cursor

AI-powered IDE with MCP integration and virtual key support

Open →

Zed Editor

High-performance editor with AI integration

Open →

Roo Code

AI-powered VS Code extension with multi-provider support

Open →

Configuration

Section titled “Configuration”

Agent integrations work with your existing DeepIntShield configuration. Ensure you have:

Monitoring Agent Traffic

Section titled “Monitoring Agent Traffic”

All agent interactions are automatically logged and can be monitored in the DeepIntShield dashboard. You can filter by provider, model, or search through conversation content to track your agents’ performance.

Agent Monitoring For complete monitoring capabilities, see Built-in Observability.

MCP Tools Integration

Section titled “MCP Tools Integration”

Connect any MCP-compatible client (Claude Code, Cursor, and other MCP-compatible applications) to DeepIntShield’s MCP server at /mcp. Its catalogue contains only tools allowed by the presented Virtual Key’s explicit MCP bindings, and each call still passes canonical authorization when GAF is enabled.

Connect Claude Code to DeepIntShield MCP

Section titled “Connect Claude Code to DeepIntShield MCP”

/mcp does not allow anonymous access. Add DeepIntShield with a valid Virtual Key using Claude Code’s JSON configuration format:

Terminal window
claude mcp add-json deepintshield '{"type":"http","url":"https://app.deepintshield.com/mcp","headers":{"Authorization":"Bearer sk-ds-your-virtual-key"}}'

Replace sk-ds-your-virtual-key with your actual DeepIntShield Virtual Key (keys are prefixed sk-ds-).

Claude Code will only have access to the specific MCP tools permitted by the virtual key’s configuration. To grant access to additional tools, verify or modify the virtual key’s MCP tool permissions in the DeepIntShield dashboard.

Supported Agents

Section titled “Supported Agents”

You can connect any MCP-compatible client to DeepIntShield’s MCP server:

  • Claude Code - Add via claude mcp add command
  • Cursor IDE - Configure in cursor settings
  • Custom Applications - Any app implementing the MCP protocol

Tool Access Control

Section titled “Tool Access Control”

Control which tools each client can access using Virtual Keys:

  • Create a Virtual Key for each client/environment
  • Configure which MCP servers and tools the key can access
  • DeepIntShield automatically enforces these permissions

For complete setup instructions and tool filtering options, see MCP Gateway URL.

Next Steps

Section titled “Next Steps”