Defenter Proxy MCP server
Provides real-time semantic security for AI coding agents by monitoring prompts, MCP tool calls, and file operations to prevent data leaks.
Also known as Defenter-AI/defenter-proxy, defenter-proxy
- Stars
- 3
- Tools
- —
- Downloads, 30 days
- 176
- Latest version
- 0.0.96
- Last commit
- Dec 22, 2025
- License
- Apache-2.0
What is Defenter Proxy MCP?
Defenter Proxy provides real-time semantic security for AI coding agents by monitoring prompts, MCP tool calls, and file operations to prevent data leaks.
Summary generated from the project's README
About Defenter Proxy MCP server
Securing AI coding agents from accidental data leaks while working in VS Code, Cursor, or Claude Code is the primary function of Defenter Proxy. It acts as a semantic policy broker that monitors prompts, MCP tool calls, and file or shell operations to ensure sensitive information stays protected. By analyzing the intent and content of agent actions rather than just the destination, it prevents the exposure of secrets or customer data during development.
The system operates through a local Python-based proxy and middleware installed on your machine, which works alongside an IDE extension. While the middleware handles client-side redaction of PII and secrets, a cloud-powered policy engine uses a classifier to evaluate minimal, redacted payloads. This setup allows the engine to return real-time decisions to allow, redact, block, or request more information without disrupting the developer workflow. Users can view all agent decisions and monitoring data directly within their IDE interface.
What you can do with it
- 1Preventing secret exposure in promptsAsk the AI to refactor code while ensuring any hardcoded credentials in the prompt are automatically redacted before being sent.
- 2Securing MCP tool interactionsMonitor and control the data being passed through MCP server calls to prevent unauthorized information transfer.
- 3Auditing agent file and shell accessTrack and validate every file read or shell command that an AI agent attempts to execute on your local system.
- 4Blocking malicious prompt injectionsUse the cloud policy engine to detect and stop prompt injection attempts in real time.
Written from the project's README. Every use case is checked against a quote from it.
How to install Defenter Proxy MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall Defenter Proxy MCP in Claude Code
$ claude mcp add defenter-proxy -- uvx defenter-proxyInstall Defenter Proxy MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"defenter-proxy": {
"command": "uvx",
"args": [
"defenter-proxy"
]
}
}
}Install Defenter Proxy MCP in VS Code
.vscode/mcp.json
{
"servers": {
"defenter-proxy": {
"type": "stdio",
"command": "uvx",
"args": [
"defenter-proxy"
]
}
}
}Install Defenter Proxy MCP in VS Code CLI
$ code --add-mcp '{"name":"defenter-proxy","type":"stdio","command":"uvx","args":["defenter-proxy"]}'Install Defenter Proxy MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"defenter-proxy": {
"command": "uvx",
"args": [
"defenter-proxy"
]
}
}
}Install Defenter Proxy MCP in Codex
$ codex mcp add defenter-proxy -- uvx defenter-proxyInstall Defenter Proxy MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.defenter-proxy] command = "uvx" args = ["defenter-proxy"]
Install Defenter Proxy MCP in Gemini CLI
$ gemini mcp add -s user defenter-proxy uvx -- defenter-proxyInstall Defenter Proxy MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"defenter-proxy": {
"command": "uvx",
"args": [
"defenter-proxy"
]
}
}
}Automated checks
Ran Nov 15, 2026Full report →Not verified yet. A server needs a score of 80, no serious finding, and enough data for us to check. How checks work · What to fix for more points
- Source repository linked and reachabledefenter-ai/defenter-proxy is public and reachablegithub.com/defenter-ai/defenter-proxy ↗Pass
- Package names this repository as its sourceNot checked
- Build provenance (Sigstore attestation)No build attestation published for this releaseNot checked
- Open source license (OSI)Apache-2.0spdx.org/licenses/Apache-2.0.html ↗Pass
- Commit or release in the last yearLast commit or release 2025-12-22github.com/defenter-ai/defenter-proxy ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNo popular package one edit away from defenter-proxyPass
- No critical advisory in the package or its dependenciesCritical advisory: GHSA-vv7q-7jx5-f767osv.dev/vulnerability/GHSA-vv7q-7jx5-f767 ↗Finding
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubDefenter
Real-time semantic security for AI coding agents and MCP tools in VS Code, Claude Code and Cursor.
Defenter monitors every prompt, every coding agent tool call, every MCP server call, and key file and shell operations in your IDE. It acts as a semantic policy broker that understands what agents are doing, not just where they send data, so you can use AI coding agents without leaking secrets or customer data.
🚀 How to use
The simplest way to use Defenter is to install the VS Code or Cursor extension:
- VS Code Marketplace: Install Defenter
- Open VSX (Cursor and others): Install Defenter
- Claude Code: install Defenter plugin
Once installed, the extension automatically intercepts and protects:
- MCP server calls
- Coding agent prompts and responses
- File reads and shell commands triggered by the agent
No manual MCP configuration is needed.
Overview
Defenter is a semantic policy broker for AI coding agents. It adds an intelligent security layer inside your IDE that:
- Intercepts every coding agent prompt and action
- Wraps every MCP tool call and response
- Analyzes the payload for sensitive information and risky behavior in real time
- Enforces your security policies with allow, redact, or block decisions
Traditional security tools cannot see what an agent is about to share or execute. They look at apps and destinations, not at the intent and content of an agent’s actions.
Defenter bridges this gap by:
- Preventing data leaks and context contamination
- Providing clear, visual monitoring of every agent decision
Architecture and how it works
This repository contains the Defenter proxy and related components that secure MCP and coding agent traffic.
Defenter is built as a Python based proxy and local middleware that the IDE extension uses to enforce policy. At a high level:
-
Local middleware layer
- Runs on the developer machine
- Hooks coding agent prompts, file reads, and shell executions
- Intercepts all MCP tool calls and responses
- Performs client side redaction of secrets and PII
-
Cloud powered policy engine
- Receives a minimal, redacted payload
- Uses a classifier and analyzer to check for data leaks, context contamination, and prompt injection
- Returns Allow, Redact, Need more info, or Block decisions in real time
- Works with low latency to make sure the development flow in without friction
-
IDE integration
- Seamless integration with VS Code and Cursor extensions
- Shows a live monitoring view of all agent actions and Defenter decisions directly inside the IDE
Together, these pieces let you harness AI coding agents and MCP tools without compromising the security of your code, data, or workflows.
References
Python Proxy: See src/README.md for detailed implementation documentation
VSC Extension: See targets/vsc-extension/README.md for installation and user guide
Defenter Proxy MCP questions
How do I add Defenter Proxy MCP to Claude Code?
Run claude mcp add defenter-proxy -- uvx defenter-proxy in your terminal. Add --scope user to make it available in every project.
How do I add Defenter Proxy MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
What does Defenter monitor?
It monitors every prompt, coding agent tool call, MCP server call, and key file and shell operations within your IDE.
How do I use Defenter with MCP?
Once the extension is installed, it automatically intercepts MCP server calls without requiring manual MCP configuration.
Which IDEs are supported?
Defenter is compatible with VS Code, Cursor, and Claude Code.
How does the security enforcement work?
The system analyzes payloads in real time and enforces security policies using allow, redact, or block decisions.