Model Context Protocol moderate risk

Currents

The Currents MCP server provides AI agents with access to test results and CI metadata for debugging and optimization, using an API key for authentication.

Connections & Capabilities

Connects To

GitHub

Capabilities

readwrite

Quickstart

Install

npx -y

Config

{
  "mcpServers": {
    "currents": {
      "command": "npx",
      "args": ["-y", "@currents/mcp"],
      "env": {
        "CURRENTS_API_KEY": "your-api-key"
      }
    }
  }
}

Exposed MCP Tools (6)

safe

currents-get-projects

Retrieves a list of projects from the Currents platform.

Read-only operation; no data modification.

safe

currents-get-run-details

Retrieves detailed information about a specific test run.

Read-only operation; no data modification.

high

currents-cancel-run

Cancels a test run that is currently in progress.

Stops the run and prevents further execution.

critical

currents-delete-run

Deletes a test run and all associated data.

Irreversible data loss.

moderate

currents-create-action

Creates a new action for a project.

Creates a new action, potentially triggering unwanted behavior.

high

currents-update-webhook

Updates an existing webhook.

Modifies existing webhook configuration, potentially disrupting integrations.

Safety Assessment

This MCP server offers valuable insights into test results but includes tools that can modify or delete data. It's relatively safe for read-only operations, but using write operations requires careful consideration of the potential impact and security of the API key.

API key authentication
Granular tools with varying danger levels
No shell execution capabilities
Clear documentation on API key security

API key grants access to test results and CI metadata
Several tools allow for destructive operations (delete, cancel, reset)
No built-in sandboxing
Potential for data exposure if API key is compromised