A self-hosted AI organization — scans your digital life, infers your goals, staffs expert teams, and autonomously drives progress through a Team Leader that manages the entire operation

The unified, OpenAI-compatible AI router. 12+ providers, 100+ models, automatic fallback. Built solo from a rented room in Peru. Star + donate to keep it alive.

JARVIS-level AI assistant with 70+ features: Voice control, Computer Vision, AR, Gesture recognition, Multi-provider LLM (6 providers), Bug bounty automation, Code generation, Business intelligence. 2-4x more powerful than ChatGPT!

Harness-Cli Agentic Development multi-provider - Claude Code Free Local

Invisible AI assistant for Windows — real-time STT + multi-provider LLM (Claude/GPT/Ollama/LM Studio) + RAG, stays invisible during screen recordings via WDA_EXCLUDEFROMCAPTURE. Electron + React + Python sidecar. Phase 1 active dev. Open to investment / partnership / acquisition.

llmgate — 高性能 Go LLM 网关，统一 20+ 供应商 API。 统一 API 接入 20+ LLM 供应商（OpenAI、Anthropic、Gemini 等） 函数调用（Tool Use）全协议支持 自动降级 + 熔断 + 重试 + 限流 成本分析 + 延迟监控 + 结构化日志 两种模式：Go SDK 直用 / 独立 HTTP 服务 内置 Web 控制台（频道管理、对话调试、Mock 测试）

Open-source non-commercial Telegram bot for type 1 diabetes — food photo → AI carb counting → bread units → diary. Self-hosted, bilingual (RU/EN). Python + litellm + Gemini Vision + aiosqlite. Active dev, contributors welcome under PolyForm Noncommercial 1.0.0.

AI-assisted structured deep thinking — pin any AI reply to branch an isolated sub-thread, infinite nesting, one-click merge. Next.js + FastAPI + Supabase + 5-provider SmartRouter.

Production n8n workflows. Hardened patterns, multi-provider LLM, no memory required. For cross-session memory see studiomeyer-io/n8n-templates.

Spatial workspace for patching AI signal graphs — wire markdown, PDFs, code, and multi-provider LLM calls into a canvas

AI agent evaluation framework for energy analytics — multi-provider LLMs, benchmark runner, MCP-based tool ecosystem.

firebat