Nervous System MCP server
Enforces deterministic governance and security linting for MCP server configurations through rule-based auditing and preflight checks.
Also known as levelsofself/mcp-nervous-system, mcp-nervous-system
- Stars
- 4
- Tools
- 2
- Downloads, 30 days
- —
- Latest version
- 1.12.1
- Last commit
- Jul 17, 2026
- License
- MIT
What is Nervous System MCP?
Nervous System enforces deterministic governance and security linting for MCP server configurations through rule-based auditing and preflight checks. It exposes 2 tools, such as audit_mcp_config, check_preflight.
Summary generated from the project's README
About Nervous System MCP server
Nervous System provides deterministic governance and security linting for MCP server configurations. It is designed for developers who need to audit their setups for security risks like inline secrets, unpinned packages, or broad filesystem scopes. The system operates through rule-based auditing and preflight checks to ensure configurations meet specific safety standards.
Users can run the audit_mcp_config tool locally using npx mcp-nervous-system to inspect files like claude_desktop_config.json. This local tool works entirely offline and returns a security score from 0 to 100 along with specific findings. For those requiring server-side capabilities, a hosted API is available via 100levelup.com, which uses a pay-per-call model with USDC on Base for governance linting and audit-chain verification.
Beyond simple linting, the check_preflight tool manages file protection by reading a protected_files_list from a configuration file or environment variable. If the list is unreadable, the system fails closed to maintain security. The package also includes several get_ tools that provide framework templates and instructions for implementing the seven core rules.
What you can do with it
- 1Local configuration security auditRun a local check on your Claude Desktop configuration to find unpinned packages or exposed secrets.
- 2Automated preflight file protectionVerify that your environment adheres to a specific list of protected files before proceeding with operations.
- 3Paid remote governance lintingUse a hosted API to perform governance linting through a pay-per-call service.
Written from the project's README. Every use case is checked against a quote from it.
How to install Nervous System MCP
Formats verified 10 Nov 2026Add to CursorInstall Nervous System MCP in Claude Code
$ claude mcp add mcp-nervous-system -- npx -y mcp-nervous-systemInstall Nervous System MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"mcp-nervous-system": {
"command": "npx",
"args": [
"-y",
"mcp-nervous-system"
]
}
}
}Install Nervous System MCP in VS Code
.vscode/mcp.json
{
"servers": {
"mcp-nervous-system": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"mcp-nervous-system"
]
}
}
}Install Nervous System MCP in VS Code CLI
$ code --add-mcp '{"name":"mcp-nervous-system","type":"stdio","command":"npx","args":["-y","mcp-nervous-system"]}'Install Nervous System MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"mcp-nervous-system": {
"command": "npx",
"args": [
"-y",
"mcp-nervous-system"
]
}
}
}Install Nervous System MCP in Codex
~/.codex/config.toml
[mcp_servers.mcp-nervous-system] command = "npx" args = ["-y", "mcp-nervous-system"]
Install Nervous System MCP in Gemini CLI
~/.gemini/settings.json
{
"mcpServers": {
"mcp-nervous-system": {
"command": "npx",
"args": [
"-y",
"mcp-nervous-system"
]
}
}
}Nervous System MCP tools 2
Detected from the READMEaudit_mcp_configcheck_preflight
| audit_mcp_config | — | |
| check_preflight | — |
README
View on GitHubNervous System
Deterministic governance lint for MCP server configurations, plus reference tooling for the 7 rules.
What each tool actually does
| Tool | Does real work? |
|---|---|
audit_mcp_config |
Yes. Deterministic lint, zero dependencies, fully offline, never throws. Checks inline secrets in env, shell wrappers, unpinned packages, auto-install flags, broad filesystem scopes, non-TLS remote transports, unpinned docker tags. Returns findings with severities and a 0-100 score. |
check_preflight |
Yes, when configured. Reads protected_files_list from nervous-system.config.json or NERVOUS_SYSTEM_PROTECTED_LIST. With no list it says so rather than guessing. If the list is unreadable it FAILS CLOSED. |
get_* (everything else) |
No. They return text: the framework, templates, instructions. That is why they are named get_. |
Install
npx mcp-nervous-system
Free local tool: audit_mcp_config
Pass your config as config (object) or config_json (string), e.g. the contents of
claude_desktop_config.json. Returns findings with severities and a 0-100 score.
Hosted API (pay-per-call, x402, USDC on Base)
The write-capable toolset runs server-side and is not in this package.
POST https://api.100levelup.com/x402/audit-mcp- governance lint ($0.05 USDC)GET https://api.100levelup.com/x402/verify-audit- audit-chain verification ($0.005 USDC)
Discovery: /openapi.json and /llms.txt.
2.0.0 breaking changes
emergency_kill_switch->get_kill_switch_instructions. It never stopped anything.verify_audit_chain->get_audit_verification_instructions. It never verified a chain.dispatch_to_llm->get_dispatch_command. It never dispatched.check_preflightno longer matches hardcoded filenames from the author's own server. It reads your list.- Hardcoded fleet counts and violation totals removed from
get_origin_story. A number baked into a package is a fossil the moment it ships.
License
MIT
Nervous System MCP questions
How do I add Nervous System MCP to Claude Code?
Run claude mcp add mcp-nervous-system -- npx -y mcp-nervous-system in your terminal. Add --scope user to make it available in every project.
How do I add Nervous System MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
What tools does Nervous System MCP provide?
It exposes 2 tools, including audit_mcp_config, check_preflight.
What does the audit_mcp_config tool do?
It performs deterministic linting to check for inline secrets, unpinned packages, broad filesystem scopes, and other security risks, returning a severity score.
How does check_preflight handle protected files?
It reads a protected files list from a configuration file or environment variable, and it fails closed if the list is unreadable.
Is there a hosted version of these tools?
Yes, there is a hosted API available for governance linting and audit-chain verification via pay-per-call using USDC on Base.