NSR MCP Server

Use NSR as a verification oracle from any MCP host — the model proposes, NSR proves.

NSR MCP Server

Exposes a running NSR API to any MCP host — Claude Code, Claude Desktop, the Agent SDK — as agent-native tools and prompts.

Why this exists

An LLM agent is a fallible neural reasoner. NSR turns a consequential call — issue this refund, grant this access, cancel this subscription — from “trust the model” into “the model proposes, NSR proves.”

The intended pattern is NSR as a verification oracle: before acting, the agent routes the decision through nsr_decide and proceeds only on an approved verdict backed by cited rules.

A refused verdict is the correct, safe result — not an error. The agent should ask for the missing facts or escalate, rather than guessing. The verify_before_acting prompt hands your host exactly this guardrail.

Configure

No npm dependencies. Node 18+ (for global fetch) and a hand-rolled JSON-RPC loop — nothing to install.

Variable	Default	Meaning
NSR_API_URL	http://127.0.0.1:8080	Base URL of the NSR server
NSR_API_KEY	—	Sent as x-api-key on every request. Required
NSR_ORG_ID	—	Optional X-Org-ID override
NSR_TIMEOUT_MS	—	Per-request timeout, so an agent never hangs on a slow endpoint

Run

# 1. A local NSR API
NSR_API_KEYS="mcp-demo-key" NSR_NEURAL_BACKEND=mock \
  ./target-local/release/nsr-server serve -p 8787

# 2. Point the MCP server at it
NSR_API_URL=http://127.0.0.1:8787 NSR_API_KEY=mcp-demo-key node nsr-mcp-server.mjs

NSR_NEURAL_BACKEND=mock is required offline. The default backend calls an external embeddings API, so entity creation returns a 500 without it.

Connect

{
  "mcpServers": {
    "nsr": {
      "command": "node",
      "args": ["/absolute/path/to/stateset-nsr/mcp/nsr-mcp-server.mjs"],
      "env": {
        "NSR_API_URL": "https://your-nsr-host",
        "NSR_API_KEY": "your-api-key"
      }
    }
  }
}

Tools

Decisions — the flagship surface:

Tool	Purpose
nsr_decide	Verified decision: approved / denied / refused with a cited proof chain
nsr_get_decision	Retrieve a prior decision by id
nsr_verify_plan	Check a multi-step plan before executing it

Teaching it rules and facts:

Tool	Purpose
nsr_add_rule	Add a rule to the org knowledge base
nsr_add_fact	Assert a fact
nsr_add_entity	Create an entity
nsr_apply_template	Apply a policy template
nsr_list_templates	List available templates

Reasoning and inspection:

Tool	Purpose
nsr_chat	Run the chat pipeline
nsr_forward_chain	Forward-chain inference
nsr_backward_chain	Backward-chain a goal
nsr_query_kb	Query the knowledge base
nsr_gss_seed_info	Grounded Symbol System seed state
nsr_flywheel_curve	Read the learning curve

Which NSR MCP server?

There are two, and they are not equivalent. The in-process Rust binary (nsr_mcp) embeds an engine and stubs the reasoning tools. Use the Node server documented here for the full, live reasoning surface.

Client-side validation

The server validates input before calling the API and names the exact malformed rule or field, rather than surfacing a cryptic 422 deserialize error — worth knowing when an agent is authoring rules and getting them wrong.

Related

• Verified Decisions API — the /v1/decisions contract these tools wrap
• Neuro-Symbolic Architecture
• All MCP servers