Currents MCP logo

Currents MCP

by currents-dev

Connects AI coding agents to Currents for CI test diagnosis, run management, metrics, Jira issues, and test actions.

Install Currents MCP

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "currents-dev-currents-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@currents/mcp"
      ],
      "env": {
        "CURRENTS_API_KEY": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

We couldn’t confirm it starts correctly

We ran the install command but it didn't respond within our test window — this can mean a slow first-time install rather than a real problem.

npx -y @currents/mcp

No response to initialize.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 1mo ago.

Overview

currents-dev/currents-mcp MCP server connects Cursor and Claude to Currents test data through MCP tools. It lets agents inspect CI runs, test failures, execution context, performance metrics, evidence artifacts, quarantine actions, webhooks, and Jira workflows. Reach for it when an agent needs real Playwright execution data to investigate failures or manage Currents resources without switching to the dashboard.

Use cases

  • Diagnose Playwright failures from CI execution data
  • Inspect flaky tests and historical test results
  • Manage quarantine, skip, and tag actions
  • Create Jira issues from failed run tests
  • Review test performance and error trends

Key features

  • Query Currents projects, runs, tests, and executions
  • Retrieve failure context and test evidence
  • Manage test actions and project webhooks
  • Analyze test, spec-file, and error metrics
  • Create or link Jira issues
  • Control and reset CI test runs

Tools43

What Currents MCP lets your agent do. Expand a tool for its parameters.

Self-reported~1.1k tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 2d ago

Recent health check succeeded.

18/18 checks healthy over the last 77d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 8d ago via OSV.dev · @currents/mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars21Stargazers on the source repository.
npm downloads87kPackage downloads in the last 30 days.
Last commit3d agoMost recent push to the default branch.
Install checkInconclusiveDidn't respond in our test window — often a slow first install.
Tools exposed43Callable tools this server registers over MCP.
Directory activity4 viewsConfig copies, upvotes, and views on AllMCPs.

How Currents MCP works

What currents-dev/currents-mcp MCP server does

The currents-dev/currents-mcp MCP server exposes Currents dashboard and CI test data to MCP-compatible coding agents. It is aimed at diagnosing Playwright test failures using recorded execution data rather than inferred results. Agents can inspect runs, spec instances, test results, failure context, error aggregates, historical performance, and test evidence such as screenshots, videos, traces, and attachments.

The server also supports operational workflows around Currents projects. Agents can create, update, enable, disable, and archive actions such as quarantine, skip, or tag rules. They can manage project webhooks, inspect pull-request cards, retrieve project terms, and create or link Jira issues through the organization’s Jira integration.

How it works

MCP clients invoke named Currents tools, with each tool handling a focused query or mutation. Read operations cover projects, runs, individual executions, test signatures, affected tests, errors, performance, and evidence. Date-range and filtering options are available for several project and execution queries.

Run operations include finding runs, retrieving run details, cancelling an in-progress run, resetting failed spec files for re-execution, deleting runs, and cancelling a run through a GitHub Actions workflow run ID and attempt number. Test debugging can be narrowed to a run, instance, or test level through failure-context retrieval.

Jira operations use the organization’s existing Jira integration. The available tools can list Jira projects and issue types, create an issue from a run test, or link an existing issue to one.

Setup and configuration

Install currents-dev/currents-mcp MCP server with the project’s documented command:

Terminal
npx -y @currents/mcp

Configure the CURRENTS_API_KEY environment variable with a Currents API key. The repository’s Cursor installation configuration passes this variable to the server.

The README identifies Cursor and Claude as supported agent integrations. It does not provide client-specific configuration details beyond the Cursor installation link and environment-variable example.

Tools and capabilities

The documented tool groups include:

  • Actions: list, create, retrieve, update, archive, enable, and disable project actions.
  • Affected tests: inspect tests and executions influenced by actions within a date range.
  • Projects and analytics: list projects, retrieve a project, inspect insights, terms, error metrics, spec-file performance, and test performance.
  • Runs and debugging: query runs, inspect details, find runs, cancel or reset runs, delete runs, inspect spec instances, retrieve historical test results, and get failure context.
  • Collaboration: list pull-request cards and create or link Jira issues.
  • Artifacts and automation: retrieve test evidence, manage webhooks, and cancel GitHub Actions CI runs.
  • Test identity: generate signatures from a project, spec-file path, and test title.

Limitations and notes

The currents-dev/currents-mcp MCP server depends on access to Currents data and a configured API key. It also does not document permissions, rate limits, retention, webhook payloads, or the exact schemas for tool arguments and responses.

Actions that change or remove data include updating, disabling, archiving, cancelling, resetting, deleting, and creating resources. Review agent permissions before enabling these operations in an automated workflow. Jira tools likewise depend on an organization Jira integration being available.

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Currents MCP

Run `npx -y @currents/mcp` and configure the `CURRENTS_API_KEY` environment variable.

It can retrieve run details, test results, failure context, execution data, performance metrics, errors, and evidence artifacts such as screenshots, videos, traces, and attachments.

Yes. It can create a Jira issue from a run test or link an existing issue through the organization’s Jira integration.

The documented configuration uses `CURRENTS_API_KEY`.

Related MCP servers

Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewCurrents MCP AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/currents-dev-currents-mcp?style=directory)](https://allmcps.com/mcp/currents-dev-currents-mcp)
HTML Embed
<a href="https://allmcps.com/mcp/currents-dev-currents-mcp"><img src="https://allmcps.com/api/badge/currents-dev-currents-mcp?style=directory" alt="Currents MCP on AllMCPs" /></a>