{"id":"raghvendra-raghuvanshi-selenium-mcp-server","name":"selenium-mcp-server","homepage":null,"repo_url":"https://github.com/Raghvendra-Raghuvanshi/selenium-mcp-server","category":"automation","subcategories":[],"tags":["mcp","selenium","browser-automation","llm","cursor","webdriver"],"what_it_does":"Provides an MCP server that lets an MCP client (e.g., Cursor) control web browsers via Selenium WebDriver, exposing tools for navigation, interaction, waiting, screenshots, and extracting page information like an accessibility snapshot.","use_cases":["LLM-assisted browser automation for QA workflows","Automated form filling and UI regression checks","Scraping page structure/links via accessibility snapshots","Generating screenshots for documentation or monitoring","CI-driven end-to-end flows using an MCP client"],"not_for":["High-security environments requiring strict browser isolation or strong auth controls","Financial/PII processing where data exfiltration risk must be minimized","Use cases requiring a public REST/HTTP API for integration (it is MCP-first)","Deterministic, fully idempotent operations without side effects"],"best_when":"You want an agent to drive a real browser with a standardized MCP tool interface and you can manage local/headless browser execution and environment setup.","avoid_when":"You need audited security boundaries, centralized auth, or you cannot control what the agent can do within the browser (e.g., arbitrary URL navigation).","alternatives":["playwright-mcp (Playwright-backed MCP server)","Selenium + a custom MCP adapter","Browser automation via headless scripts without MCP (e.g., Playwright/Selenium direct)"],"af_score":56.5,"security_score":25.2,"reliability_score":25.0,"package_type":"mcp_server","discovery_source":["github"],"priority":"low","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-04-04T21:20:02.022572+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":null,"has_sdk":false,"sdk_languages":[],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":[],"oauth":false,"scopes":false,"notes":"No authentication mechanism is described for the MCP server. This likely means access control is left to the network/process environment (e.g., local-only usage), not app-level auth."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"No pricing information provided (appears to be an open-source npm package)."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":56.5,"security_score":25.2,"reliability_score":25.0,"mcp_server_quality":70.0,"documentation_accuracy":75.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":95.0,"rate_limit_clarity":0.0,"tls_enforcement":30.0,"auth_strength":10.0,"scope_granularity":10.0,"dependency_hygiene":45.0,"secret_handling":40.0,"security_notes":"MCP server access control is not documented; without authentication/authorization, any party that can reach the process could potentially drive a browser and perform actions like visiting arbitrary URLs. The README does not describe secret handling practices, sandboxing, or safeguards against prompt-driven navigation to sensitive sites. TLS is not meaningfully applicable from the README (no mention of remote endpoints or HTTPS); security likely depends on local execution and network controls. Dependency list suggests typical JS tooling, but no vulnerability/SBOM/SCA info is provided.","uptime_documented":0.0,"version_stability":40.0,"breaking_changes_history":30.0,"error_recovery":30.0,"idempotency_support":"false","idempotency_notes":null,"pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Tool calls can have side effects in the browser (click/type/navigation), so replays are not inherently safe/idempotent.","Agent reliability may depend on dynamic pages and robust use of wait tools; the README lists waits but does not specify retry/backoff behavior.","Running with non-headless/visible mode may affect execution speed and test stability; headless/headful differences can break selectors.","Browser/environment setup (Java, browser binaries/drivers) is required; misconfiguration may surface as runtime failures rather than structured MCP errors."]}}