Sign in 中文
All features

MCP Integration

Give AI agents approved access to local backend tools

Unfour exposes approved local API, SSH, and database capabilities as MCP tools that MCP-compatible AI agents can use during backend troubleshooting.

It is for developers who want an AI agent to gather runtime evidence beyond the codebase while keeping operations inside a local, workspace-scoped tool boundary.

Read the docs

What MCP adds to the Unfour workspace

The local stdio MCP server lets an agent move from code-level reasoning to API, server, and database evidence through Unfour’s command and policy layers.

API evidence

Inspect saved requests and redacted history, review environments, and send policy-allowed requests.

SSH and runtime evidence

Review command history, run scoped diagnostics, inspect remote files, and work with saved SSH tasks.

Database evidence

List saved connections and schemas, run read-only queries or explain output, and use policy-controlled execution.

Shared workspace context

Resolve the active workspace, environment, variables, connections, and recent local activity before acting.

Policy and confirmation

Evaluate every tool call against workspace policy and require content-bound confirmation for guarded high-risk actions.

MCP client setup

Use guided setup for Codex and Cursor, or manually configure another MCP-compatible client that supports a local stdio server.

Example workflow

Assemble evidence before deciding on a fix

The agent can gather and connect evidence, but Unfour does not run an automatic troubleshooting playbook. The developer reviews the findings and decides what should change.

  1. 1. API returns 504
  2. 2. Inspect request
  3. 3. Check SSH logs
  4. 4. Query database
  5. 5. Developer decides

AI-assisted, not fully autonomous

The local workspace defines what the agent can do

MCP operations run through the local Unfour workspace. Tools are explicit, scoped, redacted where needed, and subject to policy before the agent receives evidence or performs an action.

Credentials remain stored locally by default and raw secret values are not returned by MCP tools.

Tools operate on saved, workspace-scoped API, SSH, and database resources.

Production defaults are read-only, while guarded high-risk actions require confirmation.

The AI agent gathers evidence and proposes next steps; the user remains in control.

Unfour MCP questions

Short answers about compatibility, credentials, permissions, and control.

What is Unfour MCP?

Unfour MCP is a local stdio server that exposes workspace-scoped API, SSH, database, activity, and system tools through the same command and policy layers used by the desktop application.

Which AI agents can use Unfour MCP?

Unfour provides guided configuration for Codex and Cursor. Other MCP-compatible clients can connect when they support a manually configured local stdio server.

Do SSH or database credentials leave my machine?

Credentials remain stored locally by default and are resolved on the device only when a requested operation opens a remote connection. Raw credential values are not returned to the AI agent.

Can an AI agent execute any command automatically?

No. Agents can call only the tools Unfour exposes. Calls are workspace-scoped, checked against environment policy, and guarded high-risk actions require content-bound confirmation.

Does Unfour automatically fix production issues?

No. Unfour MCP supports a diagnose-then-act workflow, but it does not run an autonomous troubleshooting playbook. The developer reviews evidence and decides the fix.

Related features

Connect your AI agent to backend evidence, not unrestricted access

Download Unfour and configure the local MCP server from the desktop application.

Download Unfour Read the docs View all features