How do I use @cqnce/mcp-server?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@@cqnce/mcp-server Request human approval before deploying the payments service to production" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

@cqnce/mcp-server

by cqnce-app

Overview Schema Related Servers Score Discussions

TypeScript

Remote

@cqnce/mcp-server

An MCP (Model Context Protocol) server that exposes cQnce human-in-the-loop authorization as AI-callable tools.

Connect it to Claude Desktop, Cursor, GitHub Copilot, or any MCP-compatible AI agent so that the agent can request human approval before performing risky or irreversible actions.

Why

AI agents increasingly perform consequential actions autonomously — deploying code, running database migrations, sending payments, modifying credentials. cQnce inserts a human into that loop: the agent asks for approval, a human approves or rejects on their phone, and the agent proceeds only if approved.

Related MCP server: Datashift MCP Server

Quick start

1. Install

npm install -g @cqnce/mcp-server
# or run without installing:
npx @cqnce/mcp-server

2. Configure environment

Variable	Required	Description
`CQNCE_BASE_URL`	✅	Base URL of your cQnce backend (e.g. `https://api.example.com`)
`CQNCE_API_KEY`	For request tools	Project API key — grants access to request submission and monitoring
`CQNCE_ADMIN_TOKEN`	For management tools	Tenant admin JWT — grants access to project, agent, team, and callback management

3. Add to Claude Desktop (`claude_desktop_config.json`)

{
  "mcpServers": {
    "cqnce": {
      "command": "npx",
      "args": ["@cqnce/mcp-server"],
      "env": {
        "CQNCE_BASE_URL": "https://api.example.com",
        "CQNCE_API_KEY": "your-project-api-key"
      }
    }
  }
}

4. Add to Cursor (`.cursor/mcp.json`)

{
  "mcpServers": {
    "cqnce": {
      "command": "npx",
      "args": ["@cqnce/mcp-server"],
      "env": {
        "CQNCE_BASE_URL": "https://api.example.com",
        "CQNCE_API_KEY": "your-project-api-key"
      }
    }
  }
}

Available tools

🔑 Human-in-the-loop approval gate

These tools use CQNCE_API_KEY.

Tool	Description
`wait_for_approval`	Primary tool. Submit a request and block until a human approves or rejects. Proceed only if `status === "APPROVED"`.
`submit_authorization_request`	Submit a request and return immediately with a `requestId`.
`poll_request_status`	Check the current status of a request.
`cancel_request`	Cancel a pending request.

Example: safe deployment

Please deploy the payments service to production.

[Agent calls wait_for_approval with payload:
  { action: "deploy", service: "payments", environment: "production", commit: "abc123" }]

[Human approves on their phone]

[Agent proceeds with deployment]

🔍 Request monitoring & audit

These tools use CQNCE_API_KEY (or CQNCE_ADMIN_TOKEN for cross-project queries).

Tool	Description
`list_requests`	List requests with optional filters (status, project, date range, tags).
`get_request`	Get full details of a single request including agent responses.
`query_request_logs`	Query the event-level audit log (routing events, timeouts, decisions).

📋 Project & routing management

These tools require CQNCE_ADMIN_TOKEN.

Tool	Description
`list_projects`	List all projects in the tenant.
`get_project`	Get project details including routing rules and callbacks.
`create_project`	Create a new project.
`update_project`	Update project settings.
`list_routing_rules`	List routing rules for a project.
`create_routing_rule`	Create a routing rule (assigns agents/teams, sets mode and timeout).
`update_routing_rule`	Update an existing routing rule.
`delete_routing_rule`	Delete a routing rule.

👥 Agent & team management

These tools require CQNCE_ADMIN_TOKEN.

Tool	Description
`list_agents`	List all agents in the tenant with online status.
`get_agent`	Get details of a specific agent.
`invite_agent`	Invite an agent by phone number.
`list_agent_invites`	List all pending/accepted/declined invitations.
`list_teams`	List all agent teams.
`create_team`	Create a new team.
`add_agent_to_team`	Add an agent to a team.
`remove_agent_from_team`	Remove an agent from a team.

📡 Webhook callback management

These tools require CQNCE_ADMIN_TOKEN.

Tool	Description
`list_callbacks`	List webhook callbacks for a project.
`create_callback`	Add a webhook endpoint (returns signing secret).
`update_callback`	Update a callback URL, events, headers, or retry count.
`delete_callback`	Remove a webhook callback.
`rotate_callback_secret`	Rotate the HMAC signing secret for a callback.

MCP resources

When CQNCE_ADMIN_TOKEN is set, the server exposes:

Resource URI	Description
`cqnce://projects`	All projects in the tenant (browsable without calling a tool).

Routing modes

When creating routing rules or submitting requests with routingMode, the supported modes are:

Mode	Behaviour
`PARALLEL`	All agents notified simultaneously; first response wins.
`SERIAL`	Agents tried one at a time in order; next notified only after timeout of current.
`MAJORITY`	Waits for a configurable quorum of matching decisions.
`CHAIN`	Every agent in the chain must respond (not just the first).

Deployment

Publishing to npm

The package must be published to npm before users can run it with npx @cqnce/mcp-server or install it globally with npm install -g @cqnce/mcp-server.

# 1. Build the TypeScript sources
cd mcp
npm run build

# 2. (Optional) Verify what will be published
npm pack --dry-run

# 3. Publish (requires npm login with publish rights to @cqnce)
npm publish

The publishConfig in package.json targets the public npm registry with scope @cqnce. Increment the version field in package.json before each publish (npm version patch|minor|major).

Running from source (no npm publish required)

If you haven't published the package yet, you can still point any MCP client at the local build by using the node command instead of npx:

{
  "mcpServers": {
    "cqnce": {
      "command": "node",
      "args": ["/absolute/path/to/cqnce/mcp/dist/index.js"],
      "env": {
        "CQNCE_BASE_URL": "https://api.example.com",
        "CQNCE_API_KEY": "your-project-api-key"
      }
    }
  }
}

Build the server first with npm run build from the mcp/ directory whenever source files change.

Security

The MCP server runs locally via stdio transport — no network port is opened.
API keys and admin tokens are read from environment variables and never logged.
The server does not store or cache credentials between invocations.
Webhook signing secrets returned by create_callback and rotate_callback_secret are shown once — store them securely.

This server cannot be installed

license - permissive license

quality - not tested

maintenance

How are these scores calculated?

Maintenance

–Maintainers

–Response time

–Release cycle

–Releases (12mo)

Commit activity

Resources

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

vantagate-mcp-server
Aderix
A
license
-
quality
D
maintenance
Human-in-the-Loop authorization gateway for AI Agents. Securely pause MCP workflows and route high-risk actions to human approvers via Slack or Email.
Last updated 2026-05-05
83
1
MIT
Datashift MCP Serverofficial
datashift-io
A
license
-
quality
D
maintenance
Enables AI agents to submit tasks for human or AI review and receive decisions via MCP tools, adding human review checkpoints to workflows.
Last updated 2026-03-19
MIT
cowork-mcp-server
CoWork-Human-Agent-Protocol
F
license
-
quality
C
maintenance
Adds trust, handoffs, and accountability to any AI agent through MCP tools for proposal approval, trust scoring, and policy enforcement.
Last updated 2026-04-24
4
AgentsGateofficial
Agent Orchestration Security
agentsgate
A
license
-
quality
B
maintenance
Enables AI agents to securely call MCP tools with risk scoring, checkpoints, rollback, and approval workflows.
Last updated 2026-07-28
MIT

View all related MCP servers

Related MCP Connectors

Oakallow
Runtime permission, approval, and audit layer for AI agent tool execution.
ToolAPI
Free public MCP for AI agents — 193 tools, 44 workflows. No API key.
PlanWright
Control plane for autonomous software labor. Agents claim objectives over MCP with audit trail.

View all MCP Connectors

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/cqnce-app/cqnce-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server

@cqnce/mcp-server

Why

Quick start

1. Install

2. Configure environment

3. Add to Claude Desktop (claude_desktop_config.json)

4. Add to Cursor (.cursor/mcp.json)

Available tools

🔑 Human-in-the-loop approval gate

Example: safe deployment

🔍 Request monitoring & audit

📋 Project & routing management

👥 Agent & team management

📡 Webhook callback management

MCP resources

Routing modes

Deployment

Publishing to npm

Running from source (no npm publish required)

Security

Maintenance

Resources

Looking for Admin?

Related MCP Servers

vantagate-mcp-server

Datashift MCP Serverofficial

cowork-mcp-server

AgentsGateofficial

Related MCP Connectors

Latest Blog Posts

MCP directory API

3. Add to Claude Desktop (`claude_desktop_config.json`)

4. Add to Cursor (`.cursor/mcp.json`)