local-ai

Wolffish is a local desktop AI agent that runs on a user’s machine with full system access. It emphasizes offline operation, Markdown-based memory/brain files, and simple skill-based extension via SKILL.md.

QVAC SDK 0.12.0 adds TurboQuant, a KV-cache quantization feature for local LLM inference that claims up to 5x more context on-device with no retraining or code changes.

A local-first AI memory engine for asking questions over selected files on your computer using Ollama, embeddings, and a custom vector database.

Harbor is a local AI stack and launcher for starting pre-configured agentic coding tools against different backends, with support for tools like Codex, OpenCode, Claude Code, and web-enabled workflows.

Local push-to-talk dictation for macOS that uses Whisper, stays on-device, and pastes text at the cursor.

Local privacy gateway that redacts sensitive text before sending prompts to external LLMs, then reconstructs the final answer locally.

A browser-based clicker game where local AI builds a custom game from your idea entirely on-device via WebGPU, with no API keys or servers.

Open protocol and reference router for capability-aware AI routing across local and cloud models, with install steps and deployment details in the README.

Didon is an AI-powered automatic time tracker for macOS that turns screen activity into daily work logs, project breakdowns, and productivity insights. It runs locally with on-device AI, supports CSV exports, and is positioned for people who want automatic journaling without manual timers.

PocketWebTools is a collection of browser-based utility tools, including AI features like background removal, image upscaling, OCR, subject cutout, and audio transcription. The site emphasizes local processing, no signup, no tracking, and data staying on device.

A small open model trained to turn English requests into Bash commands, with published training data, GGUF weights, and benchmark results.

BiNeuron is an open-source local AI assistant for code generation, file analysis, and document/OCR handling. It detects programming languages, selects models based on hardware, and can run offline with proxy support when needed.

A private, local-first browser extension for reading articles and PDFs with on-device AI summaries, text-to-speech, and local search.

A blog post describing Annex, a local-AI security tool that encrypts selected files during agent sessions to keep sensitive data away from local AI tools. The post explains the threat model, encryption flow, and the stack used to build it.

A local Go tool that turns encrypted iOS WhatsApp backups into a searchable workspace for AI agents, with OCR, transcription, and SQLite/FTS5 indexing.

Chrome extension that uses Gemini Nano via Chrome’s Prompt API to rewrite comments and posts live in a chosen tone, fully on-device with no API keys or backend.

An Emacs package that acts like an AI pair programmer and teaching assistant while you code, with save-time reviews, follow-up questions, and architecture memory for local projects.

MindShield is a browser extension that adds a reflection pause before sending simple prompts to AI tools, using local on-device classification to reduce false positives and keep prompts private.

A bootable Linux-based OS prototype with a React desktop, Rust daemons, Electron shell, and a local AI assistant. The page describes the boot flow, architecture layers, built-in apps, persistence, and hardware boot timing.

An offline agent mesh for Termux/Android that runs a self-testing, SQLite-based agent system on-device. The repo shows install steps, tier-by-tier commands, and a detailed breakdown of the mesh components and design rules.

TZRO is a local-first MCP gateway that offloads token-heavy repository and file operations to local models, aiming to reduce cloud AI token costs for coding workflows.

CoreTex is an open-source, Unix-inspired AI harness and knowledge engine that runs on flat files, with local indexing, sandboxed execution, and Obsidian-oriented workflows.

A Chrome extension that acts like a floating AI sidekick for browsing, with voice commands, page interaction, and local AI/Ollama support.

Vomit is a Go-based CLI that converts Claude output into readable English by sending it through a separate local LLM. It supports install/setup commands, a scrub mode for Claude hooks, and sidecar commands for listing and tailing sessions.