--- title: BoxLang AI 3.6: Your Agent Now Has Eyes, Ears, and a Voice description: BoxLang AI 3.6 adds browser automation, audio tools, streaming speech, and a stdio MCP client to the JVM-based BoxLang framework. Agents can now drive a real... canonical: https://daily.dev/posts/boxlang-ai-3-6-your-agent-now-has-eyes-ears-and-a-voice-kldfgxf8n twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: BoxLang AI 3.6: Your Agent Now Has Eyes, Ears, and a Voice | daily.dev og:description: BoxLang AI 3.6 adds browser automation, audio tools, streaming speech, and a stdio MCP client to the JVM-based BoxLang framework. Agents can now drive a real... og:url: https://daily.dev/posts/boxlang-ai-3-6-your-agent-now-has-eyes-ears-and-a-voice-kldfgxf8n og:image: https://api.daily.dev/og/posts/kLdfgXF8N.png og:image:alt: BoxLang AI 3.6: Your Agent Now Has Eyes, Ears, and a Voice og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # BoxLang AI 3.6: Your Agent Now Has Eyes, Ears, and a Voice **[Foojay.io](https://daily.dev/sources/foojayio)** · 10 min read · 0 upvotes · 0 comments ## Summary BoxLang AI 3.6 adds browser automation, audio tools, streaming speech, and a stdio MCP client to the JVM-based BoxLang framework. Agents can now drive a real browser through bx-playwright, use 26 bx-audio tools, and stream text-to-speech with aiSpeakStream(), supported by OpenAI, ElevenLabs, Cartesia, Mistral and Gemini. The release adds MariaDBVectorMemory for vector search on MariaDB 11.7+, raises the default timeout to 90 seconds, updates default models (gpt-5.6-luna, claude-sonnet-5), and auto-fences context to protect against indirect prompt injection. It requires BoxLang 1.18+ and Java 21+. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How does the new MCP stdio client work in BoxLang AI 3.6? It launches any MCP server as a child process and communicates over stdin and stdout, with no HTTP server or port needed. BoxLang AI handles the initialize handshake, version negotiation, and keep-alive, with one background thread reading stdout and routing JSON-RPC responses by id. A dead server restarts automatically up to a configurable maxRestarts limit. _Teams wiring agents to MCP servers can track framework updates like this on daily.dev._ ### How do I stream text-to-speech audio in BoxLang AI as it's generated? Use the new aiSpeakStream() BIF, backed by the IAiSpeechStreamService interface and speechStream provider capability. It streams audio chunk by chunk via a callback receiving typed events (audio, timestamps, done), so you can relay audio to a browser or voice interface without waiting for the full render. OpenAI, ElevenLabs, Cartesia, Mistral and Gemini support it. _Developers building real-time voice features can follow streaming speech APIs like this on daily.dev._ ### What are the minimum requirements to run BoxLang AI 3.6 with browser automation and audio tools? BoxLang AI 3.6 requires BoxLang 1.18+ and Java 21+. Optional MCP examples using npx require Node.js/npm, and examples using uvx require the uv tool. Browser automation needs the bx-playwright module installed and its drivers downloaded via 'bxplaywright install', while audio features need the bx-audio module. _Developers evaluating JVM-based agent stacks can compare setup requirements like these on daily.dev._ ## Similar posts on daily.dev - [foojay – a place for friends of OpenJDK](https://daily.dev/posts/foojay-a-place-for-friends-of-openjdk-gsr7pl0sh) · Foojay.io · 0 upvotes · 0 comments - [Join Our Webinar: Build Agents, RAG Pipelines and More](https://daily.dev/posts/join-our-webinar-build-agents-rag-pipelines-and-more-eu68yqap5) · Foojay.io · 1 upvotes · 0 comments - [BoxLang AI 3.4.0: Gateways, Human-in-the-Loop, and a Full Security Stack](https://daily.dev/posts/boxlang-ai-3-4-0-gateways-human-in-the-loop-and-a-full-security-stack-wzsdtv3ih) · Foojay.io · 0 upvotes · 0 comments - [LangSmith Agent Builder now in Public Beta](https://daily.dev/posts/langsmith-agent-builder-now-in-public-beta-gbamkfvam) · LangChain · 2 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#java](https://daily.dev/tags/java), [#mcp](https://daily.dev/tags/mcp), [#text-to-speech](https://daily.dev/tags/text-to-speech) [View this post on daily.dev](https://daily.dev/posts/boxlang-ai-3-6-your-agent-now-has-eyes-ears-and-a-voice-kldfgxf8n) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"BoxLang AI 3.6: Your Agent Now Has Eyes, Ears, and a Voice","url":"https://daily.dev/posts/boxlang-ai-3-6-your-agent-now-has-eyes-ears-and-a-voice-kldfgxf8n","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/boxlang-ai-3-6-your-agent-now-has-eyes-ears-and-a-voice-kldfgxf8n"},"datePublished":"2026-10-31T01:18:54.270Z","dateModified":"2026-10-31T01:19:22.875Z","description":"BoxLang AI 3.6 adds browser automation, audio tools, streaming speech, and a stdio MCP client to the JVM-based BoxLang framework. Agents can now drive a real...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e07ac3ae7de82e9c328937e997170e9a?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e07ac3ae7de82e9c328937e997170e9a?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Foojay.io","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Foojay.io","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/1820f6a6ae944760aa322b67ed85f848","url":"https://daily.dev/sources/foojayio"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/boxlang-ai-3-6-your-agent-now-has-eyes-ears-and-a-voice-kldfgxf8n","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,java,mcp,text-to-speech","timeRequired":"PT10M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Foojay.io","item":"https://daily.dev/sources/foojayio"},{"@type":"ListItem","position":3,"name":"BoxLang AI 3.6: Your Agent Now Has Eyes, Ears, and a Voice"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/boxlang-ai-3-6-your-agent-now-has-eyes-ears-and-a-voice-kldfgxf8n#faq","mainEntity":[{"@type":"Question","name":"How does the new MCP stdio client work in BoxLang AI 3.6?","acceptedAnswer":{"@type":"Answer","text":"It launches any MCP server as a child process and communicates over stdin and stdout, with no HTTP server or port needed. BoxLang AI handles the initialize handshake, version negotiation, and keep-alive, with one background thread reading stdout and routing JSON-RPC responses by id. A dead server restarts automatically up to a configurable maxRestarts limit. Teams wiring agents to MCP servers can track framework updates like this on daily.dev."}},{"@type":"Question","name":"How do I stream text-to-speech audio in BoxLang AI as it's generated?","acceptedAnswer":{"@type":"Answer","text":"Use the new aiSpeakStream() BIF, backed by the IAiSpeechStreamService interface and speechStream provider capability. It streams audio chunk by chunk via a callback receiving typed events (audio, timestamps, done), so you can relay audio to a browser or voice interface without waiting for the full render. OpenAI, ElevenLabs, Cartesia, Mistral and Gemini support it. Developers building real-time voice features can follow streaming speech APIs like this on daily.dev."}},{"@type":"Question","name":"What are the minimum requirements to run BoxLang AI 3.6 with browser automation and audio tools?","acceptedAnswer":{"@type":"Answer","text":"BoxLang AI 3.6 requires BoxLang 1.18+ and Java 21+. Optional MCP examples using npx require Node.js/npm, and examples using uvx require the uv tool. Browser automation needs the bx-playwright module installed and its drivers downloaded via 'bxplaywright install', while audio features need the bx-audio module. Developers evaluating JVM-based agent stacks can compare setup requirements like these on daily.dev."}}]} ```