Linebreak Gate MCP server
{
"mcpServers": {
"linebreakapp-linebreak-gate": {
"command": "uvx",
"args": [
"linebreak-gate"
],
"env": {
"LINEBREAK_LICENSE_KEY": "<linebreak_license_key>",
"ANTHROPIC_API_KEY": "sk-ant-..."
}
}
}
}{
"mcpServers": {
"linebreakapp-linebreak-gate": {
"command": "uvx",
"args": [
"linebreak-gate"
],
"env": {
"LINEBREAK_LICENSE_KEY": "<linebreak_license_key>",
"ANTHROPIC_API_KEY": "sk-ant-..."
}
}
}
}{
"mcpServers": {
"linebreakapp-linebreak-gate": {
"command": "uvx",
"args": [
"linebreak-gate"
],
"env": {
"LINEBREAK_LICENSE_KEY": "<linebreak_license_key>",
"ANTHROPIC_API_KEY": "sk-ant-..."
}
}
}
}{
"servers": {
"linebreakapp-linebreak-gate": {
"command": "uvx",
"args": [
"linebreak-gate"
],
"env": {
"LINEBREAK_LICENSE_KEY": "${input:linebreak_license_key}",
"ANTHROPIC_API_KEY": "${input:anthropic_api_key}"
}
}
},
"inputs": [
{
"id": "linebreak_license_key",
"type": "promptString",
"description": "Linebreak license key",
"password": true
},
{
"id": "anthropic_api_key",
"type": "promptString",
"description": "Anthropic api key",
"password": true
}
]
}{
"mcpServers": {
"linebreakapp-linebreak-gate": {
"command": "uvx",
"args": [
"linebreak-gate"
],
"env": {
"LINEBREAK_LICENSE_KEY": "<linebreak_license_key>",
"ANTHROPIC_API_KEY": "sk-ant-..."
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall uvx linebreak-gateversion 1.10.3
What it does
Linebreak Gate is a security enforcement tool that runs CVE scans and optional AI code reviews at the git/CI boundary, serving approved acceptance criteria over MCP to Claude Code, Cursor, or Codex. It blocks merges on vulnerable dependencies and first-party code issues, requiring human sign-off for fixes or overrides.
- Block pull requests that introduce known CVEs in dependencies, with dependency scanning across npm, PyPI, Go, Cargo, Maven and other ecosystems.
- Run AI-powered static analysis on first-party source code to detect injection, broken auth, secret exposure, SSRF, unsafe deserialization, and crypto misuse (Pro feature).
- Enforce approved acceptance criteria in pull requests via MCP integration with Claude Code, Cursor, or Codex, requiring human review and recorded approvals before merge.
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.