#local-llm

Secure AI Engineering Framework 2026: Data-Boundary Security for Frontier Models

fak — the Fused Agent Kernel: one Go binary that turns a tool-using agent (Claude Code, Codex, Cursor, any OpenAI/Anthropic/MCP client) into a managed agent: cache-stable model traffic, context compaction + crash resume, nanosecond tool-call policy, local GGUF serving with SSD expert offload.

Fast-ASDLC: 5x TTM with AI-native Agentic SDLC. Local-LLM first, Human-in-the-loop, Spec-driven. Built on DDD, Hexagonal Architecture, C4 Model & MCP. Features Meta-agents for self-improvement, Memory Bank for context persistence, and automated 100% test coverage. Everything-as-Code & Mermaid.js centric to save context window and slash token costs.

Experimental Chrome side-panel browser agent with LangGraph.js, MCP/WebMCP, local models, and semantic browser automation.

A quantum prompt playground & macOS-native Rust coding agent: morphing fluid entropy into rigid execution checkpoints. Entropy cannot lie.

Your own self-hosted AI agent OS

caix — native Apple Core AI inference server for Apple silicon (beta): OpenAI/Anthropic API, dashboard, streaming chat with tools/skills/MCP, MTP speculative decoding.

Desktop AI chat app unifying local Ollama models with OpenAI and Claude — includes an agentic mode where the model can read/write files and run shell commands, with every action requiring your approval.

An AI that stays with you between conversations. Keeps the threads you didn't finish, answers from your own notes with receipts you can open, and asks before it acts on your behalf. Provider-neutral: any model, local or cloud, your choice.

A privacy-first, self-hosted AI assistant with smart model routing and local LLM support (Ollama).

🏭 Coding agent CLI engineered to make any LLM — local or cloud, frontier or 7B — actually useful. 16 providers, multi-tab sessions, automatic key+model failover, plan-mode review, MCP, and recovery loops for models that misbehave.

A sandboxed, full-featured agent operating system in a single Go binary. Bring any model, manage local models, point it at a goal, and grant it real authority: every action is sandboxed, governed, and sealed into a verifiable, tamper-evident record an independent party can check. Runs interactive or 24/7, or embed it in your own system.

Open-source Astro MDX tech blog for AI setup guides, local LLMs, Ollama, developer tools, self-hosted tutorials, and automation workflows

A local multi-model gateway for Codex & Claude Code on Apple Silicon — a Qwable-first agent runtime (planner → executor → repair → critic → judge), fully offline.

Docker Compose stack for serving a local, OpenAI-compatible LLM (vLLM on Intel XPU) on an Intel Arc Pro B60 GPU — reproducible config with operator and developer docs.

One-Click Private Agentic Coding Launch Tool with Local llama.cpp for Developer AI in 2026

AgentTrail is a local-first AI agent layer for Ollama/local models that shows its work: semantic search, diff-safe edits, receipts, replay, memory, MCP, reports, and trust controls.

2026 Ultimate Podman AI Coding Agent – Zero-Trace, MIT Licensed GitHub Repo

AI-powered TRPG engine — LLM as GM, streaming narrative, dice system, screenplay-driven gameplay. Vue 3 + FastAPI. 本地 LLM 驱动的 AI 跑团引擎。大语言模型担任 GM，支持流式叙事、骰子判定、剧本驱动推进。Vue 3 + FastAPI。

Local-first desktop agent that compiles AI trajectories into deterministic, invariant-gated automations. No code required. Explore once with a model. Replay forever without one.

Talunor is a local-first terminal AI agent written in Go, with persistent, auditable SQLite memory, guarded tool use, and a complete step-by-step course explaining how it is built.

Local-first sovereign Life OS with Obsidian-compatible vault, durable AI agents, local retrieval, and Cogni-Brain integration.

An AI leaderboard scoring chatbots on how much they bullshit vs. give honest answers to unanswerable questions. Works with paid APIs and free/local models.

Local-first threat detection agent for Windows. Runs a 4B-parameter LLM (Qwen3.5-4B) on-device via KoboldCpp. ReAct FSM, MITRE ATT&CK, YARA, OSINT, 15 skills, 4000+ tests.

Chiron Rising — an Alpha Centauri (SMACX) mod pack: LLM-written faction diplomacy with leader memory, generated base names, Planetnet dispatches, probe protests, three new factions, and optional Future Society repricing. Built on Thinker.

VSCode + VSCodium extension for Ollama Modelfiles. Syntax highlighting, 18-rule linter (unknown PARAMETERs, missing FROM, bad MESSAGE roles, quoting errors), hovers, autocomplete, and 26 snippets. Works in vscode.dev, Cursor, and Windsurf.

Run local coding agents with full tool access and browser-based monitoring using an uncensored Ollama model on your hardware.

A local-first coding agent: runs the heavy executor on your local model and escalates to the cloud only when stuck. Out-resolves single-shot Opus/Sonnet by planning, running the project's real tests, and retrying — at ~half the cost of an all-cloud chain. Graded by SWE-bench, not an LLM.

幻语AI (HuanyuAI Chat) - 基于 Android WebView 的 AI 聊天客户端开源版

A fully local, zero-cost RAG chatbot for mac.du.ac.in — built with FAISS, Llama 3.2 3B, FastAPI, and React. CPU-only, 8GB RAM, zero cloud dependency.

The 48GB VRAM Local LLM Guide: real benchmarks, model configs, and gotchas from 135+ community reports. Includes runnable benchmark scripts.

Claude Code plugin: delegate small tasks (commit messages, PR descriptions, shell/code drafts, log digests) to local Ollama models - private data stays on your machine

Portable local-first AI agent harness - local LLMs via Ollama, tool-calling with tiered permissions, Telegram remote. Tauri 2 + React.

Claude Code plugin that delegates simple, well-scoped tasks to a locally running LLM (Ollama, LM Studio, llama.cpp, vLLM) via MCP

Benchmark local coding LLMs on an OpenAI-compatible endpoint, then keep the serving config that won. Hidden executable tests, reference-validated tasks, mermaid reports.

Run Qwen3.8-27B fully offline on an Apple Silicon Mac (MLX, 4/5/8-bit) and use it as the local backend for Claude Code. Uncensored/abliterated build supported. No API key, no network. Built by yempik.

AI-generated git commit messages in VS Code — local-first: Ollama and any OpenAI-compatible endpoint. Conventional Commits + gitmoji, offline, no telemetry.

On/off switch for routing Claude Code through a local model instead of Anthropic Pro/Max subscription. No API keys, no cloud fallback—local mode either works or fails cleanly. (tldr; Route Claude Code through a local model via Distrobox, powered by llama.cpp and LiteLLM.)

Digital Garage Enterprise

Archives local LLMs for long-term offline use — runtime-independent copies of weights, tokenizers, licenses, and checksums, with offline smoke tests.