What runtime or dependencies are required?
The server runs via `npx -y @currents/mcp`, which requires Node.js and npm (typically bundled with Node). No additional installation beyond the MCP client setup is needed.
Currents MCP Server
@currents-dev
基本情報
Currents MCP Server について
Currents MCP Server
Currents MCP Server の設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"currents": {
"command": "npx",
"args": [
"-y",
"@currents/mcp"
],
"env": {
"CURRENTS_API_KEY": "your-api-key"
}
}
}
}Currents MCP Server のツール
39List all actions for a project with optional filtering. Actions are rules that automatically modify test behavior (skip, quarantine, tag). Supports filtering by status (active/disabled/archived/expired) and search by name. Requires a projectId.
Create a new action for a project. Actions define rules that automatically skip, quarantine, or tag tests based on conditions like test title, file path, git branch, etc. Requires projectId, name, action array, and matcher object.
Get a single action by ID. The actionId is globally unique, so projectId is not required. Returns full action details including matcher conditions and current status.
Update an existing action. The actionId is globally unique. You can update name, description, action array, matcher, or expiration date. All fields are optional.
Delete (archive) an action. This is a soft delete - the action will be marked as archived but not permanently removed. The actionId is globally unique.
Enable a disabled action. Changes the action status from disabled to active, making it apply to matching tests again. The actionId is globally unique.
Disable an active action. Changes the action status to disabled, temporarily preventing it from applying to tests. The actionId is globally unique.
List tests affected by actions (quarantine, skip, tag) for a project within a date range. Returns aggregated data grouped by test signature. Supports filtering by action types, action ID, status, and search. Requires projectId, date_start, and date_end. Preview endpoint: fields and path may change.
Get execution details for a specific affected test (by signature) within a date range. Returns individual test execution records with action info. Uses cursor-based pagination. Requires projectId, signature, date_start, and date_end.
List test executions where a specific action/rule was applied, within a date range. Uses cursor-based pagination. Requires actionId, date_start, and date_end.
Retrieves projects available in the Currents platform. Supports cursor-based pagination with limit, starting_after, ending_before parameters, or set fetchAll=true for automatic pagination. This is a prerequisite for using any other tools that require project-specific information.
Get a single project by ID. Returns project details including name, creation date, failFast setting, inactivity timeout, and default branch name.
Get aggregated run and test metrics for a project within a date range. Returns overall metrics and timeline data with configurable resolution (1h/1d/1w). Supports filtering by tags, branches, groups, and authors. Requires projectId, date_start, and date_end.
List pull-request cards for a project (runs grouped by meta.pr.id). Supports cursor pagination, runs_per_pr preview count, and filters by tags, branches, authors, and latest-run status. Requires projectId.
List cursor-paginated project terms for one type (tag, branch, authorName, etc.). Supports search, sort direction, and starting_after or ending_before cursors. Requires projectId and termType.
Create a Jira issue from a run test using the organization Jira integration. Requires projectId, runId, testId, jiraInstallationId, jiraProjectId, and jiraIssueType. Optional customFields array.
Link an existing Jira issue to a run test using the organization Jira integration. Requires projectId, jiraIssueKey, runId, testId, jiraInstallationId, jiraProjectId, and jiraIssueType. Optional comment and includeContextInComment.
List Jira projects available for the organization integration. Use returned project IDs as jiraProjectId when creating issues. Requires jira_installation_id.
List Jira issue types and custom fields for a Jira project. Requires jiraProjectId and jira_installation_id.
Retrieves a list of runs for a specific project with optional filtering. Supports filtering by branch, tags (with AND/OR operators), status (PASSED/FAILED/RUNNING/FAILING), completion state, date range, commit author, and search by ciBuildId or commit message. Requires a projectId. If the projectId is not known, first call 'currents-get-projects' and ask the user to select the project.
Retrieves details of a specific test run. Requires a user-provided runId.
Find a run by query parameters. Returns the most recent completed run matching the criteria. Can search by ciBuildId (exact match) or by branch/tags. Supports pwLastRun flag for Playwright last run info. Requires projectId.
Cancel a run in progress. This will stop the run and mark it as cancelled. Requires runId.
Reset failed spec files in a run to allow re-execution. Requires runId and machineId array (1-63 machine IDs). Optionally supports batched orchestration.
Delete a run and all associated data. This is a permanent deletion. Requires runId.
Cancel a run by GitHub Actions workflow run ID and attempt number. Optionally scope by projectId or ciBuildId. Requires githubRunId and githubRunAttempt.
Retrieves debugging data from a specific execution of a test spec file by instanceId.
Retrieves spec files performance metrics for a specific project within a date range. Supports ordering by avgDuration, failedExecutions, failureRate, flakeRate, flakyExecutions, fullyReported, overallExecutions, suiteSize, timeoutExecutions, or timeoutRate. Supports filtering by tags, branches, groups, and authors. Requires a projectId. If the projectId is not known, first call 'currents-get-projects' and ask the user to select the project.
Retrieves aggregated test metrics for a specific project within a date range. Supports ordering by failures, passes, flakiness, duration, executions, title, and various delta metrics. Supports filtering by spec name, test title, tags, branches, groups, authors, minimum executions, test state, and annotations. Requires a projectId. If the projectId is not known, first call 'currents-get-projects' and ask the user to select the project.
Generates a unique test signature based on project, spec file path, and test title. The test title can be a string or array of strings (for nested describe blocks). Requires a projectId. If the projectId is not known, first call 'currents-get-projects' and ask the user to select the project.
Retrieves historical test execution results for a specific test signature. Supports filtering by date range, branch, tags, git author, test status (passed/failed/pending/skipped), run group, flaky status, and annotations. Requires the test signature. If the signature is not known, first call 'currents-get-tests-signatures'.
Get test failure context for AI debugging at run, instance, or test level. Supports json or md format, detail level, and pagination for failed tests. Requires run_id for run-level, or instance_id with optional test_id.
Get aggregated error metrics for a project within a date range. Supports filtering by error_target, error_message, error_category, error_action, tags, branches, authors, and groups. Supports grouping by target, action, category, or message. Returns error counts, affected tests and branches, with timeline data. Requires projectId, date_start, and date_end.
Collect evidence artifacts (screenshots, videos, traces, attachments) produced by tests in a CI run, with signed download URLs grouped per test. Use to gather proof or a demo of an implemented feature from CI — e.g. before/after screenshots, text output stored as test attachments, or Playwright videos and traces — instead of running tests locally. Locates the run by runId, or by projectId with ciBuildId or branch (latest run). Supports filtering by spec file, test title, and test status. URLs are signed and time-limited, so download the files promptly.
List all webhooks for a project. Webhooks allow you to receive HTTP POST notifications when certain events occur in your test runs: RUN_FINISH (run completed), RUN_START (run started), RUN_TIMEOUT (run timed out), RUN_CANCELED (run was cancelled). Requires a projectId.
Create a new webhook for a project. Specify the URL to receive POST notifications, optional custom headers (as JSON string), events to trigger on (RUN_FINISH, RUN_START, RUN_TIMEOUT, RUN_CANCELED), and an optional label. Requires projectId and url.
Get a single webhook by ID. The hookId is a UUID. Returns full webhook details including url, headers, events, label, and timestamps.
Update an existing webhook. You can update the url, headers (as JSON string), hookEvents array, or label. All fields are optional. The hookId is a UUID.
Delete a webhook. This permanently removes the webhook. The hookId is a UUID.
概要
What is Currents MCP Server?
The Currents MCP Server connects AI coding agents (like Cursor and Claude) directly to the Currents dashboard, giving them full visibility into CI test results so they can diagnose flaky tests, pinpoint failures, and act on real execution data without leaving the editor.
How to use Currents MCP Server?
Obtain a Currents API key, then configure the server in your MCP client. For Cursor, add the @currents/mcp command and CURRENTS_API_KEY environment variable to mcp.json. For Claude Desktop, edit claude_desktop_config.json similarly. For Claude Code, run claude mcp add with the same parameters.
Key features of Currents MCP Server
- Query runs, spec files, and individual test results from CI.
- Surface error trends and performance metrics across your test suite.
- Manage quarantine rules, webhooks, and project settings programmatically.
- Create, link, and list Jira issues using the organization Jira integration.
- Cancel, reset, or delete runs directly from the MCP tools.
- Get aggregated error metrics and failure context for AI debugging.
Use cases of Currents MCP Server
- Diagnose and fix flaky tests by examining actual CI output through an AI agent.
- Programmatically manage project configuration (webhooks, actions, rules) from the editor.
- Automatically create Jira issues from failing tests to track technical debt.
- Monitor test suite health by querying aggregated performance metrics.
- Allow agents to cancel or reset stuck runs without switching to the Currents dashboard.
よくある質問
How is data authenticated and secured?
Authentication uses a `CURRENTS_API_KEY` environment variable. By connecting AI tools via MCP to Currents, you grant them access to your API key, test results, and CI metadata. It is your responsibility to vet any AI agents or services you use.
Where does test execution data live?
All test execution data resides in the Currents cloud platform. The MCP server queries the Currents API to retrieve it; no local storage of test results is described.
What transport protocol does it use?
The MCP server uses the stdio transport, as shown by the `command` and `args` configuration in all setup examples.
Can the server create or delete project resources?
Yes. Tools include creating, updating, and deleting actions and webhooks, as well as deleting runs and cancelling or resetting them.
「その他」の他のコンテンツ
YOUB
Connect your AI assistant to your own YOUB endurance training account. Read recorded activities with full analysis, planned sessions, race goals, wellness (HRV, sleep, resting HR) and your athletic profile. Create or mov
MaxKB
1Panel-dev🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。
22.0K
Reactive Resume
amruthpillaiA one-of-a-kind resume builder that keeps your privacy in mind. Completely secure, customizable, portable, open-source and free forever. Try it out today!
39.5K
P
PickFu MCP
PickFu MCP is a Model Context Protocol (MCP) server that connects AI assistants, developers, marketers, product managers, and AI agents to PickFu's consumer research platform. It enables users to create surveys, conduct
Maestro
mobile-dev-incPainless E2E Automation for Mobile and Web
14.8K
コメント