AegisMCP
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@AegisMCPshow the recent security alerts"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
š”ļø AegisMCP: Zero-Trust Security & Cost Firewall for AI Agents
AegisMCP is a local-first security gateway and visual dashboard that sits between your filesystem and your AI coding agents (Claude Desktop, Cursor, Copilot). It strictly protects your infrastructure from destructive commands and heavily defends your API unit economics from runaway token costs.
šØ The Problem
Giving AI agents raw, unmonitored access to the Model Context Protocol (MCP) or your local terminal is a massive risk. A hallucination or infinite loop can wipe out a database, expose your .env files, or rack up thousands of dollars in API bills overnight, destroying your project's profit margins.
⨠How AegisMCP Solves It
AegisMCP acts as an impenetrable, zero-trust proxy:
Financial Guardrails (Guaranteed Margins): Hard-caps daily API spend and enforces a strict structural markup (e.g., 30%) calculation before execution to guarantee positive unit economics.
Semantic Threat Interception: Blocks prompt injections and malicious commands (
chmod 777,DROP TABLE) instantly.Real-Time Visual Dashboard: A Next.js hub that lets you monitor agent activity and requires explicit approval for dangerous actions.
Zero-Latency Local Compute: The interceptor runs locally alongside your agent, meaning zero cloud latency and no third-party data tracking.
š Seamless Setup & Connection Guide
AegisMCP is designed for frictionless integration. You can boot the entire firewall in under 60 seconds.
1. Install & Boot the Firewall
Initialize the monorepo locally without complex configurations:
git clone [https://github.com/thekartikeyamishra/AegisMCP.git](https://github.com/thekartikeyamishra/AegisMCP.git)
cd AegisMCP
2. Connect Claude Desktop
Tell Claude to route its tools through your local Aegis firewall.
Open your Claude configuration file:
Mac: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Add AegisMCP to the mcpServers object:
npm install
npm run dev
{
"mcpServers": {
"AegisFirewall": {
"command": "node",
"args": ["/absolute/path/to/AegisMCP/packages/cli/dist/index.js"]
}
}
}
Restart Claude Desktop. You will see a šØ icon indicating Aegis is connected.
3. Connect Cursor AI
Open Cursor Settings > Features > MCP Servers.
Click + Add New MCP Server.
Name it AegisFirewall.
Set the command to node and the argument to the absolute path of packages/cli/dist/index.js.
Click Save. Cursor will now route all system calls through Aegis.
š Search Keywords & SEO Tags
mcp-server, model-context-protocol, claude-desktop-skills, ai-agent-security, llm-firewall, prompt-injection-protection, cursor-agent-tools, openai-cost-control, api-token-tracker, zero-trust-ai, local-mcp-proxy.
š¤ Contributing
AegisMCP is a community-driven defense layer. We actively accept Pull Requests! If you discover a new AI jailbreak or destructive pattern, please submit a PR. Check out our Good First Issues in the issue tracker to get started.
āļø License & Liability
Distributed under the MIT License. AegisMCP is provided "AS IS". The authors hold no liability for API costs incurred or data lost by third-party AI agents. See LICENSE for more information.This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- Alicense-qualityDmaintenanceA local-first security system for autonomous AI agents that provides tools for security verification, goal anchoring, and action logging. It protects against prompt injection and goal drift by enforcing user-defined rules and offering performance insights through session grading.Last updated18MIT
- Alicense-qualityCmaintenanceA security interceptor for AI agents that mitigates prompt injections and data loss in autonomous workflows.Last updated3211MIT
- AlicenseAqualityAmaintenanceLocal zero-trust permission gateway for AI agents. Enforces policy-based tool authorization, human approvals, scoped permissions, and cryptographically verifiable audit logs.Last updated45Apache 2.0
- AlicenseBqualityBmaintenanceWAF for AI agents ā block prompt injection before it reaches the LLM.Last updated5MIT
Related MCP Connectors
The WAF for agents. Pattern-based + heuristic firewall scans prompts, RAG documents, tool argume...
Security firewall for AI agents ā scans MCP calls for injection, secrets, and risks.
Responsible-AI guardrails for agents: scoring with policy, injection & PII detection, DPDP.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/thekartikeyamishra/AegisMCP'
If you have feedback or need assistance with the MCP directory API, please join our Discord server