Elevenlabs.io announced a new group of AI generated services, for a fee of course. This is important because it shows that AI gen is getting cheaper and cheaper for SaaS companies to offer "unrelated" services to their core value proposition. The problem is that it blurs the lines between services. That's a fast track to commoditization where one is just as good as another. The fallout could be that they go out of business leaving customers scrambling for alternatives. Outages happen with growing frequency. Vulnerabilities continue to plague the industry. It might be prudent to have a self-hosted alternative or at least alternative nodes setup and ready to go. Cheers

Anthropic's official docs now make it crystal clear: > OAuth authentication (used with Free, Pro, and Max plans) is intended exclusively for Claude Code and Claude.ai. Using OAuth tokens obtained through Claude Free, Pro, or Max accounts in any other product, tool, or service, including the Agent SDK, is not permitted and constitutes a violation of the Consumer Terms of Service. > Developers building products or services that interact with Claude's capabilities, including those using the Agent SDK, should use API key authentication through Claude Console or a supported cloud provider. Anthropic also states they reserve the right to enforce this without prior notice. If you're using OpenClaw or any similar third-party tool that routes through your Pro/Max credentials instead of proper API keys, be aware this violates the ToS and could result in enforcement action on your account. Source: [Claude Code Legal and Compliance](https://docs.anthropic.com/en/docs/claude-code/legal-and-compliance)

Code update including line counts of source and tests. Studio Lines of Code (2026-02-10) Backend — API source ............... 66,949 lines, 315 files Backend — Worker source ......... 7,098 lines, 34 files Backend source (combined) ...... 74,047 lines, 349 files ─────────────────────────────── Frontend (UI) source ................... 75,544 lines, 338 files ─────────────────────────────── Backend — API tests ................... 89,046 lines, 316 files Backend — Worker tests ............. 4,810 lines, 33 files Backend tests (combined) ......... 93,856 lines, 349 files ─────────────────────────────── Frontend — Jest tests ................. 49,036 lines, 187 files Frontend — Playwright (e2e) ...... 3,326 lines, 22 files Frontend tests (combined) ......... 52,362 lines, 209 files ════════════════════════════════════════════ TOTAL .......................................... 295,809 lines, 1,245 files ───────────── Test Cases ──────────── Backend — API unit ..................... 4,673 Backend — API integration .......... 1,269 Backend — Worker ....................... 429 Backend total ................................ 6,371 Frontend — Jest ............................ 3,718 Frontend — Playwright (e2e) ........ 231 Frontend total ................................ 3,949 ─────────────────────────────── TOTAL TEST CASES ...................... 10,320 Backend test-to-code ratio ............ 1.3:1 Frontend test-to-code ratio ............ 0.7:1

Thank you for coming to Self-Host Hub, a community for finding solutions to AI Automation. I recommend you start by browsing Start Here! Then come back and tell us about yourself, your goals, your business and how self-hosting can help.

Welcome to Self-Host Hub. This community is for people building automation infrastructure they own — self-hosted workflows, local AI, and systems you run for yourself or your clients. 📋 ROADMAP (Weekly) Directional focus areas, not delivery guarantees. - 2/9/26 Faceless video workflow end-to-end: script generation → image generation → audio → video assembly, TTS provider integration (ElevenLabs, OpenAI TTS) - 2/16/26 Debug mode: inline step editing, input/output inspection, continue/skip/stop - 2/23/26 Alpha release 📅 UPDATES - 2/8/26 JSON string extraction fix completed. Faceless video pipeline in progress. Deployment testing underway. - 1/22/26 Direction locked: deterministic workflows, structured AI, production-safe execution. - 1/10/26 Community focus narrowed to real deployments and operational tradeoffs. (Updates appended as work progresses.) WHAT IS STUDIO Self-hosted workflow automation platform for AI content pipelines. You own the infrastructure, pick your providers, keep your keys. Custom workflow engine — no n8n, no Temporal, no off-the-shelf runtime. Full control over execution, debugging, and step-level data flow. Four pieces: - App (API + web UI) - Workers (general, video/FFmpeg, ComfyUI/GPU) - Catalog (installable provider/service packages) - Marketplace (templates + premium token-gated packages) Deploy however you want: single Docker container, Docker Compose, Kubernetes, RunPod, VPS + Cloudflare Tunnel, or hybrid GPU setups over Tailscale. ✅ DONE - Custom workflow engine with step-by-step execution, pause/resume, approval gates - Multi-tenant org system with role-based access - Provider/service architecture — install only what you need - 10+ provider integrations (Claude, OpenAI, Gemini, Google, Slack, Notion, Airtable, telegram, trallo, typeform, ComfyUI, JSON2Video, etc.) - ComfyUI GPU worker with model management - Video worker (FFmpeg transcoding, Whisper STT) - Real-time WebSocket updates during workflow runs - 7 Docker images covering every deployment scenario - Marketplace backend with token-gated access - Full documentation suite - 2/1/26 Deployment testing across all 7 Docker scenarios (single container, Compose, K8s, RunPod, VPS + Cloudflare Tunnel, hybrid GPU, all-in-one)