Systems and Architecture Guides
Claude Code Without Anthropic: ANTHROPIC_BASE_URL Proxy Setup & DeepSeek-V4 Pro Routing
Run Claude Code CLI against DeepSeek-V4 Pro and Qwen 3.8 Max without an Anthropic subscription or $50/mo API bills.
The Closed-Loop Code Review & Auto-Fix Engine: Independent Second-LLM Verification
How /v1/review inspects diffs with an independent reviewer LLM (temperature 0.1), detects bugs, and feeds fix prompts back into the coder.
LockManager: Protecting Critical Files from Accidental AI Overwrites
How to lock database configs, sensitive auth routes, and core components from AI modifications during massive refactors.
AI Prompt Enhancer: Transforming Vague Prompts into Deterministic Technical Specs
How the /api/enhancer endpoint expands brief user ideas into comprehensive architectural requirements.
In-Browser WebContainers: POSIX Node.js Runtime in WebAssembly (<250ms Boot)
How Unchained Code mounts a client-side operating system inside your browser tab with zero server compute costs.
DeepSeek-V4 Pro & Qwen 3.8 Max: High-Throughput Multi-Model Pool Architecture
Multi-tier routing across DeepSeek-V4 Pro, Qwen 3.8 Max, Qwen3-Coder, Gemini 3.7 Flash, and Mistral Large.
MCP Protocol Integration: Connecting Custom External Tools & APIs
How Unchained Code supports Model Context Protocol (MCP) servers via /api/mcp-check and /api/mcp-update-config.
WebContainer Diagnostics & Memory Doctor: Preventing 4GB Wasm Heap Crashes
How /api/system/diagnostics and /api/system/disk-info keep client-side Node.js builds fast and crash-free.
GitLab & GitHub Enterprise Sync: Connecting Private Self-Hosted Repositories
How /api/gitlab-projects and /api/gitlab-branches bridge sovereign corporate git instances with zero data leaks.
Zero Data Retention & 1-Click Telemetry Opt-Out: 100% Air-Gapped Privacy
How to enforce absolute privacy, disable tracking, and ensure your source code is never stored or used for model training.
Bring Your Own Key (BYOK) & Air-Gapped Local Inference via Ollama / vLLM
Route Unchained Code through your own OpenRouter, DeepSeek-V4 Pro, or 100% private local GPUs.
The $E_u$ Equity Ledger & Dynamic Throttling: How Ad-Funded Free AI Coding Works
Deep dive into real-time token unit economics, Green/Orange/Red tiers, and sponsor impression credits.
Prompt Caching & Prefix Optimization: Slashing Latency and Token Costs by 80%
How stable system prompt architecture and KV-cache hits accelerate multi-turn developer workflows.
Fast-Lane Pro: Zero Latency, Zero Ads & High-Priority Dedicated Gateway
Unrestricted developer infrastructure for professional indie hackers, agencies, and engineering teams.
Unchained Code for macOS: Native Tauri Wrapper with Direct Local Filesystem Access
How the native macOS desktop app bridges local compiler tools, native terminals, and open-weight models.
One-Click Deploy: GitHub Git Data API, Vercel, Netlify & Supabase Integration
Export complete production codebases directly from browser memory without vendor lock-in.
E2B Firecracker Cloud Sandboxes: 150ms Cold-Start MicroVMs for Heavy Compute
Seamlessly offload tasks exceeding in-browser Wasm limits (native C/C++, Docker, headless browsers).
Bounties & Viral Referral Engine: Earn Free $E_u$ Credits and Build for Life
How open-source contributions, GitHub stars, and peer referrals fund unlimited Green-tier AI coding.