agentguard

TensorScholar/agentguard

★ 0 stars Python AI/LLM Updated 7d ago

Security gateway and audit layer for AI agents, MCP servers, and tool-calling workflows.

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "agentguard": {
      "command": "uvx",
      "args": [
        "agentguard"
      ]
    }
  }
}

Or install with pip: pip install agentguard

README Excerpt

AgentGuard is a local-first security gateway for AI agents, MCP servers, and tool-calling workflows. It answers three practical questions: 1. What tools can this agent access? 2. What dangerous actions should be blocked before execution? 3. What audit evidence exists after an agent run? AgentGuard is intentionally not an eval platform, not an observability dashboard, and not an

agentguard

Quick Install

README Excerpt

Related AI/LLM Servers