Nervous System
You'd reach for this when you need structured oversight across multiple AI agents with different roles and risk profiles. It enforces behavioral boundaries through manifest files that define permissions, escalation rules, and reliability controls per agent. The thirteen included manifests cover everything from public-facing coaching (lily) to strict financial workflows (harry) to legal guidance with compliance boundaries (aram). The drift detection catches when agents deviate from their defined parameters. If you're orchestrating specialized agents that need different trust levels and operating constraints, this gives you a governance layer instead of hoping prompt engineering holds up under load.
- GitHub stars
- 4
- tools
- 6
- transport
- STDIO
- last updated
- Jul 13, 2026
Tools
6 toolsPublic tool metadata for what this MCP can expose to an agent.
audit_mcp_configDeterministic lint for MCP server configurations that returns findings with severities and a 0-100 score.
Deterministic lint for MCP server configurations that returns findings with severities and a 0-100 score.
No parameter schema in public metadata yet.
check_preflightReads a protected files list from the config and fails closed if the list is unreadable.
Reads a protected files list from the config and fails closed if the list is unreadable.
No parameter schema in public metadata yet.
get_kill_switch_instructionsReturns text instructions for the kill switch; it does not stop anything.
Returns text instructions for the kill switch; it does not stop anything.
No parameter schema in public metadata yet.
get_audit_verification_instructionsReturns text instructions for audit verification; it does not verify a chain.
Returns text instructions for audit verification; it does not verify a chain.
No parameter schema in public metadata yet.
get_dispatch_commandReturns a dispatch command as text; it does not dispatch.
Returns a dispatch command as text; it does not dispatch.
No parameter schema in public metadata yet.
get_origin_story
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
Nervous 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)
Details
- Repository
- github.com/levelsofself/mcp-nervous-system
- Category
- AI & LLM Tools
- Package
mcp-nervous-system- Transport
- STDIO
- Registry
- MCP registry, active
More from levelsofself
Related AI & LLM Tools MCP servers
View all
Mif Toolsvarun29ankus/mif-specConvert, validate, and inspect AI agent memories across formats0stars
Vetojigyasudham/veto93 agentic MCP tools + 49 specialist agents for every major AI CLI. Self-learning, no API keys.4stars
signet-evaljmcentire/signet-evalDeterministic policy enforcement and MCP management for AI agent tool calls.4stars
Fcopjoinwell52-ai/fcopFCoP Core adapter with 25 canonical tools and 6 read-only resources for durable agent work.4stars
Penfield Memorypenfieldlabs/penfield-mcpPersistent memory and knowledge graphs for AI agents. Hybrid search, context checkpoints, and more.4stars
Tessera MCP Servertessera-llm/mcp-serverMCP server for Tessera's LLM cost-optimization layer: anchored spend, drift, recommendations.0stars