{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"npm-currents-mcp","slug":"currents-mcp","name":"@currents/mcp","type":"mcp","url":"https://currents.dev","page_url":"https://unfragile.ai/currents-mcp","categories":["mcp-servers"],"tags":["currents","mcp-server","playwright","dashboard","reporting"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"npm-currents-mcp__cap_0","uri":"capability://tool.use.integration.playwright.test.execution.orchestration.via.mcp","name":"playwright test execution orchestration via mcp","description":"Exposes Playwright test execution as MCP tools, allowing Claude and other LLM clients to invoke browser automation workflows through a standardized tool-calling interface. Implements a schema-based function registry that maps Playwright operations (navigation, interaction, assertion) to callable MCP resources with structured input/output contracts, enabling LLMs to compose multi-step browser automation sequences without direct SDK knowledge.","intents":["I want Claude to run my Playwright tests and interpret failures in natural language","I need an LLM to automate browser interactions as part of a larger workflow","I want to expose my test suite as callable tools for AI agents without writing custom API wrappers"],"best_for":["QA engineers building AI-assisted test automation workflows","Teams integrating Playwright tests into LLM-powered agents","Developers prototyping browser automation agents with Claude"],"limitations":["Requires Playwright to be installed and configured in the same environment as the MCP server","No built-in test result persistence — results are ephemeral unless explicitly captured by the client","Limited to Playwright's supported browsers (Chromium, Firefox, WebKit); no support for legacy IE or mobile-specific browsers","Execution is synchronous per tool call — parallel test execution requires client-side orchestration"],"requires":["Node.js 18+","Playwright 1.40+","@currents/mcp npm package","MCP-compatible client (Claude Desktop, custom LLM application)"],"input_types":["JSON-serialized Playwright action objects (goto, click, fill, etc.)","URL strings","CSS/XPath selectors","Text content for assertions"],"output_types":["JSON test result objects with pass/fail status","Screenshot data (base64 or file path)","Assertion error messages","Browser console logs and network activity"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-currents-mcp__cap_1","uri":"capability://tool.use.integration.currents.test.result.aggregation.and.reporting.via.mcp.tools","name":"currents test result aggregation and reporting via mcp tools","description":"Exposes Currents cloud test reporting platform as MCP callable tools, enabling LLM clients to query test runs, retrieve failure summaries, and access CI/CD test metadata without direct API calls. Implements a schema-based wrapper around Currents' REST API that translates test result queries into structured MCP tool calls, with built-in filtering, pagination, and result formatting for LLM consumption.","intents":["I want Claude to fetch the latest test failures from my Currents dashboard and summarize them","I need an AI agent to query test run history and identify flaky tests","I want to expose my CI/CD test results to an LLM for automated root-cause analysis"],"best_for":["DevOps engineers building AI-assisted test failure analysis","QA teams integrating Currents reporting into LLM workflows","CI/CD pipeline owners automating test result interpretation"],"limitations":["Requires valid Currents API credentials and active Currents account","Read-only access to test results — cannot modify or re-run tests through MCP tools","Pagination limits may require multiple tool calls for large test suites (1000+ tests per run)","Test result freshness depends on Currents' data pipeline; real-time results may lag by 30-60 seconds"],"requires":["Currents account with API access enabled","Currents API key (stored as environment variable or passed to MCP server)","Node.js 18+","@currents/mcp npm package"],"input_types":["Test run IDs (UUID format)","Project identifiers","Filter criteria (status, duration, tags)","Pagination parameters (limit, offset)"],"output_types":["JSON test run summaries with pass/fail counts","Structured test failure objects with stack traces","Test execution timelines and performance metrics","Flakiness indicators and historical trend data"],"categories":["tool-use-integration","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-currents-mcp__cap_2","uri":"capability://tool.use.integration.mcp.server.lifecycle.management.and.tool.registration","name":"mcp server lifecycle management and tool registration","description":"Implements the Model Context Protocol server specification, handling client connection negotiation, tool schema registration, and request routing. Uses a declarative tool definition system where each Playwright or Currents operation is registered as an MCP tool with JSON Schema validation, enabling clients to discover available capabilities and invoke them with type-safe parameters.","intents":["I want to connect Claude Desktop to my Playwright tests and Currents dashboard","I need to expose my test automation capabilities as discoverable MCP tools","I want to build a custom LLM client that can invoke Playwright and Currents operations"],"best_for":["Developers integrating Currents MCP into Claude Desktop or custom LLM applications","Teams building internal AI-assisted testing platforms","Organizations standardizing on MCP for test automation tooling"],"limitations":["MCP protocol overhead adds ~50-100ms per tool invocation compared to direct SDK calls","Tool discovery and schema validation happens at connection time; schema changes require server restart","No built-in authentication beyond environment variables — requires external secret management for production","Single-threaded request handling by default; high-concurrency scenarios require custom deployment configuration"],"requires":["Node.js 18+","@currents/mcp npm package","MCP-compatible client (Claude Desktop 0.1.0+, or custom implementation)","Environment variables for API credentials (CURRENTS_API_KEY, etc.)"],"input_types":["MCP protocol messages (JSON-RPC 2.0 format)","Tool invocation requests with typed parameters"],"output_types":["MCP tool schema definitions (JSON Schema format)","Tool execution results (JSON-serialized)","Error responses with diagnostic information"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-currents-mcp__cap_3","uri":"capability://image.visual.playwright.screenshot.capture.and.visual.assertion.support","name":"playwright screenshot capture and visual assertion support","description":"Enables Playwright test execution to capture screenshots and expose them as base64-encoded data or file references through MCP tools, allowing LLMs to perform visual assertions and analyze UI state. Integrates with Playwright's screenshot API to capture full-page, element-specific, or viewport-only images, with optional comparison against baseline images for regression detection.","intents":["I want Claude to take a screenshot of a webpage and describe what it sees","I need an AI agent to detect visual regressions by comparing screenshots","I want to capture UI state during test execution for debugging and documentation"],"best_for":["QA engineers building visual regression testing workflows","Teams using Claude for UI testing and accessibility analysis","Developers debugging browser automation failures with visual context"],"limitations":["Screenshot data is base64-encoded, increasing payload size by ~33% compared to binary format","Full-page screenshots may exceed LLM context windows for large, complex pages (>10MB uncompressed)","Visual comparison requires baseline images to be pre-captured and stored; no automatic baseline generation","Screenshot timing is synchronous; capturing multiple screenshots in sequence adds latency"],"requires":["Playwright 1.40+","Node.js 18+","Sufficient disk space for baseline image storage (if using visual regression)"],"input_types":["Screenshot options (fullPage, viewport, element selector)","Comparison baseline image paths","Image format specifications (PNG, JPEG)"],"output_types":["Base64-encoded image data","Image file paths","Visual comparison results (match/mismatch with diff data)","Pixel-level difference metrics"],"categories":["image-visual","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-currents-mcp__cap_4","uri":"capability://data.processing.analysis.test.failure.context.enrichment.and.error.message.extraction","name":"test failure context enrichment and error message extraction","description":"Automatically extracts and structures error messages, stack traces, and browser console logs from failed Playwright tests, enriching them with contextual metadata (test name, duration, browser type) for LLM consumption. Implements a parsing layer that normalizes error output across different assertion libraries (Playwright's built-in assertions, Chai, Jest) and formats them as structured JSON for easier LLM interpretation.","intents":["I want Claude to analyze a test failure and suggest fixes based on the error message","I need an AI agent to categorize test failures by root cause (assertion, timeout, network)","I want to automatically generate bug reports from test failures with full context"],"best_for":["QA engineers using AI to automate root-cause analysis of test failures","Teams building intelligent test failure dashboards","Developers integrating test failure analysis into CI/CD pipelines"],"limitations":["Error message parsing is heuristic-based; complex or custom error formats may not be fully extracted","Stack trace normalization assumes standard Node.js error format; custom error classes may not parse correctly","Browser console logs are captured only if explicitly enabled in Playwright configuration","Large error messages (>10KB) may be truncated to fit LLM context windows"],"requires":["Playwright 1.40+","Node.js 18+","Test execution through Currents MCP tools or direct Playwright invocation"],"input_types":["Playwright test failure objects","Error messages and stack traces","Browser console output"],"output_types":["Structured failure objects with categorized error types","Normalized stack traces with file paths and line numbers","Extracted assertion messages and expected vs actual values","Contextual metadata (test duration, browser type, retry count)"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-currents-mcp__cap_5","uri":"capability://memory.knowledge.browser.context.and.session.management.for.stateful.test.workflows","name":"browser context and session management for stateful test workflows","description":"Manages Playwright browser contexts and sessions across multiple MCP tool invocations, enabling stateful test workflows where subsequent operations inherit browser state (cookies, local storage, authentication) from previous steps. Implements a context registry that persists browser instances and page objects between tool calls, allowing LLMs to compose multi-step workflows without re-initializing the browser for each step.","intents":["I want Claude to log into a website once and then perform multiple actions in the same session","I need an AI agent to maintain browser state across multiple test steps","I want to avoid re-authenticating for each browser automation step"],"best_for":["QA engineers building complex, multi-step test workflows","Teams automating user journeys that require persistent authentication","Developers building stateful browser automation agents"],"limitations":["Context registry is in-memory; browser sessions are lost if the MCP server restarts","No built-in context serialization; cannot persist browser state to disk for recovery","Concurrent contexts may consume significant memory for large test suites (1 context ≈ 50-100MB)","Context cleanup is manual; orphaned contexts must be explicitly closed to free resources"],"requires":["Playwright 1.40+","Node.js 18+","Sufficient memory for multiple browser contexts (minimum 2GB for 10+ concurrent contexts)"],"input_types":["Context identifiers (UUID or string keys)","Browser launch options (headless, viewport, etc.)","Page navigation and interaction commands"],"output_types":["Context creation confirmations with identifiers","Page state snapshots (cookies, local storage, session storage)","Navigation and interaction results"],"categories":["memory-knowledge","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-currents-mcp__cap_6","uri":"capability://data.processing.analysis.currents.ci.cd.metadata.extraction.and.test.run.correlation","name":"currents ci/cd metadata extraction and test run correlation","description":"Queries Currents API to retrieve CI/CD metadata associated with test runs (commit hash, branch, build ID, author), enabling LLMs to correlate test failures with code changes and build context. Implements a metadata enrichment layer that combines test result data with Git and CI/CD information, providing LLMs with full context for root-cause analysis and impact assessment.","intents":["I want Claude to identify which code changes caused a test failure","I need an AI agent to correlate test failures with specific commits or pull requests","I want to automatically generate failure reports that include Git and CI/CD context"],"best_for":["DevOps engineers building AI-assisted failure analysis","Teams automating root-cause analysis across code and test data","CI/CD pipeline owners integrating test results with code change tracking"],"limitations":["Requires Currents to be integrated with your Git provider and CI/CD system","Metadata availability depends on CI/CD system configuration; some fields may be missing or incomplete","Commit and author information is read-only; cannot modify Git history through MCP tools","Large test runs (1000+ tests) may require multiple API calls to retrieve full metadata"],"requires":["Currents account with CI/CD integration enabled","Currents API key with read access","Git repository integration configured in Currents","CI/CD system integration (GitHub Actions, GitLab CI, Jenkins, etc.)"],"input_types":["Test run IDs","Commit hashes or branch names","Build IDs from CI/CD system"],"output_types":["Enriched test result objects with Git metadata","Commit information (hash, author, message, timestamp)","Build metadata (build ID, status, duration)","Pull request information (if available)"],"categories":["data-processing-analysis","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-currents-mcp__cap_7","uri":"capability://data.processing.analysis.flaky.test.detection.and.historical.trend.analysis","name":"flaky test detection and historical trend analysis","description":"Analyzes historical test execution data from Currents to identify flaky tests (tests that fail intermittently) and track failure trends over time. Implements statistical analysis of test pass/fail rates across multiple runs, with configurable thresholds for flakiness detection and trend visualization data for LLM interpretation.","intents":["I want Claude to identify which tests are flaky and suggest fixes","I need an AI agent to track test failure trends and alert on regressions","I want to automatically quarantine flaky tests and prioritize fixing them"],"best_for":["QA engineers managing large test suites with flakiness issues","Teams building intelligent test failure dashboards","DevOps engineers automating test quality monitoring"],"limitations":["Flakiness detection requires historical data; new tests have insufficient history for reliable detection","Statistical analysis assumes independent test executions; correlated failures (e.g., shared infrastructure issues) may be misclassified","Trend analysis is limited to available historical data in Currents; data retention policies may limit lookback window","Flakiness thresholds are configurable but require tuning for different test suites and environments"],"requires":["Currents account with historical test data (minimum 10-20 runs per test)","Currents API key with read access","Node.js 18+"],"input_types":["Test names or patterns","Historical lookback window (days)","Flakiness threshold (percentage)"],"output_types":["Flakiness scores (0-100, where 100 is completely flaky)","Pass/fail rate statistics","Trend data (improving, degrading, stable)","Failure frequency and patterns"],"categories":["data-processing-analysis","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":39,"verified":false,"data_access_risk":"high","permissions":["Node.js 18+","Playwright 1.40+","@currents/mcp npm package","MCP-compatible client (Claude Desktop, custom LLM application)","Currents account with API access enabled","Currents API key (stored as environment variable or passed to MCP server)","MCP-compatible client (Claude Desktop 0.1.0+, or custom implementation)","Environment variables for API credentials (CURRENTS_API_KEY, etc.)","Sufficient disk space for baseline image storage (if using visual regression)","Test execution through Currents MCP tools or direct Playwright invocation"],"failure_modes":["Requires Playwright to be installed and configured in the same environment as the MCP server","No built-in test result persistence — results are ephemeral unless explicitly captured by the client","Limited to Playwright's supported browsers (Chromium, Firefox, WebKit); no support for legacy IE or mobile-specific browsers","Execution is synchronous per tool call — parallel test execution requires client-side orchestration","Requires valid Currents API credentials and active Currents account","Read-only access to test results — cannot modify or re-run tests through MCP tools","Pagination limits may require multiple tool calls for large test suites (1000+ tests per run)","Test result freshness depends on Currents' data pipeline; real-time results may lag by 30-60 seconds","MCP protocol overhead adds ~50-100ms per tool invocation compared to direct SDK calls","Tool discovery and schema validation happens at connection time; schema changes require server restart","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.49742201635108657,"quality":0.26,"ecosystem":0.55,"match_graph":0.25,"freshness":0.52,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.15,"match_graph":0.23,"freshness":0.12}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:23.902Z","last_scraped_at":"2026-05-03T14:04:47.472Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":15349,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=currents-mcp","compare_url":"https://unfragile.ai/compare?artifact=currents-mcp"}},"signature":"Y5KrzTMJumvHRj9zwA/im/PI8hcdOGHK/mGfYEokWuaOiwdpfBedMXBzYKpVuHAw1tZwPEvKr6DHmD1NxHYiCQ==","signedAt":"2026-06-23T07:56:43.729Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/currents-mcp","artifact":"https://unfragile.ai/currents-mcp","verify":"https://unfragile.ai/api/v1/verify?slug=currents-mcp","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}