Skip to main content
Glama

vrymdor-node

The Python client for the Vrymdor AI Node by Vrymnox Systems LLC. Every call goes through your node's checks: the agent's identity and status, your policy, removal of PII (personally identifiable information) and secrets, a signed request and a metadata-only audit record. One module, one dependency (httpx).

Install

pip install vrymdor-node                 # the client
pip install "vrymdor-node[langchain]"    # plus LangChain tools
pip install "vrymdor-node[llamaindex]"   # plus LlamaIndex tools

Related MCP server: HallucC — Hallucination Fact-Check & AI Safety Guard

Use

You need your node's address and an agent registered in the Vrymdor app (AI agents, Register an agent). The agent secret is shown once.

from vrymnox_node import VrymnoxNode

node = VrymnoxNode("http://vrymdor-node:8080", agent_id="agt_...", agent_secret="vrym_agent_...")

node.sanitize("Customer jane@acme.io called")         # "Customer [REDACTED:EMAIL] called"
node.tool("github", "create_issue", {"title": "Login fails"})
node.llm("anthropic", "v1/messages", {"model": "...", "messages": [...]})

tools = node.langchain_tools()     # one LangChain StructuredTool per tool your policy allows
tools = node.llamaindex_tools()    # the same, as LlamaIndex FunctionTools

A refusal raises VrymnoxError with a stable code (for example tool_not_allowed, rate_limited, agent_stopped) and retryable. An action your policy holds raises VrymnoxHeld (code held_by_policy): it did not run, and a held action is never released. Change the policy to allow it.

Documentation: https://docs.vrymdor.com/ai-node.html

MCP: Claude Desktop, Cursor and VS Code, no Docker

The package includes vrymdor-mcp, the bridge desktop MCP clients start as a local command. It relays each MCP message to your node and renews the agent's token by itself. The easiest way to run it is uvx, which also fetches a matching Python for you: install uv once (macOS and Linux: curl -LsSf https://astral.sh/uv/install.sh | sh; Windows PowerShell: powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex").

Test the agent first (it prints "Connected: ..." or the reason it is not):

VRYMNOX_NODE_URL=https://node.vrymdor.com/t/abc123def456 VRYMNOX_AGENT_ID=agt_... \
  VRYMNOX_AGENT_SECRET=vrym_agent_... uvx --from vrymdor-node vrymdor-mcp --check

Claude Desktop (claude_desktop_config.json) and Cursor (mcp.json):

{
  "mcpServers": {
    "vrymdor": {
      "command": "uvx",
      "args": ["--from", "vrymdor-node", "vrymdor-mcp"],
      "env": {
        "VRYMNOX_NODE_URL": "https://node.vrymdor.com/t/abc123def456",
        "VRYMNOX_AGENT_ID": "agt_...",
        "VRYMNOX_AGENT_SECRET": "vrym_agent_..."
      }
    }
  }
}

VS Code (.vscode/mcp.json): the same block under "servers", with "type": "stdio" added.

If the client says it cannot find uvx (Claude Desktop on macOS starts commands without your shell's PATH), put the full path in command: the output of which uvx (macOS, Linux) or where uvx (Windows).

License

MIT. The Vrymdor AI Node itself is licensed separately under your Vrymnox agreement.

License and terms

Copyright (c) 2026 Vrymnox Systems LLC. This client is open source under the MIT License (LICENSE). Using the Vrymdor AI Node service it connects to is governed by the Vrymdor Master Subscription Agreement (https://vrymdor.com/msa) and the Acceptable Use Policy in the agreement set you accept when you sign up.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Governed MCP gateway that lets AI agents call tools with policy enforcement, prompt-injection screening, a kill-switch, and tamper-evident signed audit logs.
      Apache 2.0
    • A
      license
      A
      quality
      B
      maintenance
      HallucC — an MCP server for hallucination detection and AI agent safety. Claim-by-claim fact-checking of LLM/agent output with cited sources, a prompt-injection & jailbreak guard, and L0–L3 risk gating for computer-use (CUA) agent actions. 5 tools, 3 free checks/day, works with Claude Code, Cursor and Claude Desktop.
      5
      49 npm
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Sits between an agent and an upstream MCP server to enforce per-profile tool grants, deterministic allow/deny/handoff rules, field and PII masking, idempotent writes, and one-line audit logging from a single YAML config. Enables regulated deployments to add guardrails that fail closed without changing the customer's tools or the agent's prompt.
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables agents to route MCP tool calls through an inline policy proxy that enforces OWASP ASI safeguards, gates destructive or high-cost actions behind expiring HMAC-signed human approvals, and records tamper-evident Merkle audit receipts.
      MIT