{"id":"webhiveos-webhive","name":"WebHive","homepage":"https://webhiveos.xyz/","repo_url":"https://github.com/webhiveos/WebHive","category":"automation","subcategories":[],"tags":["ai-ml","browser-automation","agents","python","self-hosted","playwright","vnc","computer-use","llm-integration"],"what_it_does":"WebHive appears to be an AI-driven browser automation/control center. It runs a local web UI and a controllable browser session (optionally reusing an existing Chrome profile) to let an AI model perform tasks in the browser while showing interactions via a built-in VNC/noVNC viewer. It supports multiple LLM providers via API keys configured in an .env file.","use_cases":["Automating multi-step web tasks via an AI model (form filling, navigation, browsing-based research)","Running agent-like browser workflows locally with access to a persistent browser session","Watching or debugging AI-driven browser interactions using VNC/noVNC video/streaming","Using existing Chrome/Firefox/Edge profile logins for authenticated browsing"],"not_for":["Production-grade, headless, multi-tenant automation service without additional hardening","Use cases requiring fine-grained API-level access control and auditable service-to-service permissions","Tasks that require strict compliance boundaries without data-handling documentation"],"best_when":"You want local, self-hosted AI browser automation with optional persistent browser state for interactive browsing.","avoid_when":"You cannot provide LLM API keys or you need a documented public API/contract for programmatic integration; also avoid when you cannot control exposure of browser profile data and session credentials.","alternatives":["Browser-use/web-ui (referenced as a base project in the README)","Playwright-based custom automations (build your own agent loop)","Selenium/Playwright + LLM orchestration frameworks (LangChain/LlamaIndex/agent frameworks)"],"af_score":28.5,"security_score":26.0,"reliability_score":15.0,"package_type":"mcp_server","discovery_source":["github"],"priority":"low","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-04-04T19:33:23.327033+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":false,"mcp_server_url":null,"has_sdk":false,"sdk_languages":[],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":["LLM provider API keys via .env (OPENAI_API_KEY, ANTHROPIC_API_KEY, GOOGLE_API_KEY)"],"oauth":false,"scopes":false,"notes":"Authentication is primarily via upstream model API keys configured in the local environment. The README does not describe any authentication mechanism for the local web UI itself (e.g., login/auth middleware) or for accessing the VNC/noVNC endpoints."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"Project pricing is not described in the provided content; LLM usage cost depends on configured providers and usage."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":28.5,"security_score":26.0,"reliability_score":15.0,"mcp_server_quality":0.0,"documentation_accuracy":45.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":80.0,"rate_limit_clarity":0.0,"tls_enforcement":20.0,"auth_strength":30.0,"scope_granularity":0.0,"dependency_hygiene":30.0,"secret_handling":50.0,"security_notes":"The system is configured via environment variables for upstream LLM API keys, which is a good baseline. However, there is no information about transport security for the local web UI (README shows http://localhost), no described authentication/authorization for the UI/VNC endpoints, and no documented guidance on securely handling or minimizing browser profile credentials. If persistent sessions are enabled, sensitive cookies/logins may be retained on disk and accessible to the automation environment.","uptime_documented":0.0,"version_stability":25.0,"breaking_changes_history":20.0,"error_recovery":15.0,"idempotency_support":"false","idempotency_notes":null,"pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["No documented API/MCP interface for agents; integration would likely be via running the app and automating the UI/browser workflow rather than calling a stable programmatic contract.","If using persistent sessions and existing browser profiles, agent actions may depend on prior state (logins, cookies) and can cause unintended side effects.","VNC password defaults and browser-accessible UI endpoints may be exposed if the host is not properly firewalled (no guidance shown here)."]}}