Nitrowatch

v1.0.0

A powerful MCP server built with NitroStack

Connection Setup

Add via Cursor Settings UI (Settings > Features > MCP > Add New MCP Server): 

{
  "mcpServers": {
    // your other mcp servers
    "nitrowatch": {
      "url": "https://nitrowatch-6a6d3fcd-powerx-srmist.app.nitrocloud.ai/mcp"
    }
  }
}

Connect remote tools directly via Claude's Web UI:

Add custom connector BETA
Connect Claude to your data and tools. Learn more about connectors or get started with pre-built ones.
Advanced settings
Only use connectors from developers you trust. Anthropic does not control which tools developers make available and cannot verify that they will work as intended or that they won't change.
Building an MCP server? Report issues and subscribe to updates here

Configure custom tools directly via ChatGPT's Web UI:

New App
Icon (optional)
PNG only. Best results at 256 x 256 px or larger. Max file size: 10 KB
Name
Description (optional)
Connection
Authentication
Custom MCP servers introduce risk. Learn more
OpenAI hasn't reviewed this MCP server. Attackers may attempt to steal your data or trick the model into taking unintended actions, including destroying data.
Read the guide

Add the following configuration block under mcpServers in your Antigravity configuration file (~/.gemini/config/mcp_config.json): 

{
  "mcpServers": {
    // your other mcp servers
    "nitrowatch": {
      "serverUrl": "https://nitrowatch-6a6d3fcd-powerx-srmist.app.nitrocloud.ai/mcp"
    }
  }
}

Add the following configuration block to your Codex configuration file (~/.codex/config.toml): 

[mcp_servers.nitrowatch]
url = "https://nitrowatch-6a6d3fcd-powerx-srmist.app.nitrocloud.ai/mcp"

Connect directly using the Server-Sent Events endpoint: 

https://nitrowatch-6a6d3fcd-powerx-srmist.app.nitrocloud.ai/mcp
Available Tools
register_server

Register an MCP server with NitroWatch so it can be watched

discover_capabilities

Connect to a registered server's real MCP endpoint and list its tools/resources/prompts

get_burn_rate

Project time-to-exhaustion on the team's 5M AI token budget

generate_glue

Generate a stub connector function that calls a tool on one registered server and pipes its result into a tool on another

classify_tools

Assign a risk tier (read / reversible / irreversible) to every tool on a registered server. Run this after discover_capabilities. Tries the LLM classifier first and falls back to the deterministic one.

set_tier

Manually override a tool's risk tier when the classifier gets it wrong. Recorded as a human decision in the audit trail.

request_action

Ask NitroWatch to run a tool on a governed server. Read tools and promoted tools run immediately; everything else is queued for human approval and returns an approvalId.

list_pending_approvals

Show every action currently blocked and waiting on a human decision, newest first. Renders the interactive approval console.

approve_action

Human approves a pending action. Executes it, records the approval, and reports whether the tool has now earned promotion.

deny_action

Human denies a pending action. Nothing executes, and the denial is recorded permanently against that tool.

promote_tool

Grant a reversible tool standing permission to run unattended. Requires a clean approval record. Irreversible tools are refused.

revoke_tool

Withdraw a tool's autonomy immediately. It returns to requiring approval on every call.

get_trust_status

Show the approval record and autonomy state for every governed tool — the evidence behind each promotion decision.

get_audit_log

Return the full decision history: what was classified, allowed, blocked, approved, denied, promoted, and by whom.