# NeuralCleave

- [Introduction](https://docs.neuralcleave.com/introduction.md): NeuralCleave is your personal AI assistant — built on the AI stack, not bolted onto it.
- [Quickstart](https://docs.neuralcleave.com/quickstart.md): Get NeuralCleave running and answering messages in under five minutes.
- [Installation](https://docs.neuralcleave.com/installation.md): Install NeuralCleave on Linux, macOS, or Windows.
- [Configuration](https://docs.neuralcleave.com/configuration.md): Complete config.toml reference — agent, models, memory, voice, gateway, and secret resolution.
- [Architecture](https://docs.neuralcleave.com/architecture.md): How the NeuralCleave gateway, pipeline, and channel adapters fit together.
- [Memory System](https://docs.neuralcleave.com/memory.md): NeuralCleave's 3-tier memory (Redis, Qdrant, SQLite) retrieves relevant context for every query.
- [Skills](https://docs.neuralcleave.com/skills.md): Skills are Python modules NeuralCleave can write, review, and hot-load at runtime — the assistant extends itself, with a human in the loop.
- [Skills Gallery](https://docs.neuralcleave.com/skills-gallery.md): Community-built skills for NeuralCleave — copy, install, and extend your assistant.
- [LLM Providers](https://docs.neuralcleave.com/llm-providers.md): NeuralCleave supports 19 AI providers switchable at runtime without a restart.
- [Channels](https://docs.neuralcleave.com/channels.md): Connect NeuralCleave to any of 32 messaging platforms with a single config block.
- [Voice Overview](https://docs.neuralcleave.com/voice/overview.md): NeuralCleave's local voice pipeline — wake word, VAD, STT, and TTS, all in-process.
- [Wake Word](https://docs.neuralcleave.com/voice/wake-word.md): Always-on wake word detection using OpenWakeWord — cross-platform, runs 100% locally.
- [Push-to-Talk](https://docs.neuralcleave.com/voice/push-to-talk.md): Manual-trigger voice input — hold the PTT button to record, release to transcribe and reply.
- [Voice Providers](https://docs.neuralcleave.com/voice/providers.md): Configure STT and TTS backends — from local faster-whisper to ElevenLabs cloud.
- [CLI Reference](https://docs.neuralcleave.com/cli.md): All neuralcleave command-line commands.
- [Observability](https://docs.neuralcleave.com/observability.md): Metrics, structured logs, and the Observability dashboard for monitoring NeuralCleave in production.
- [Deployment](https://docs.neuralcleave.com/deployment.md): Run NeuralCleave as a systemd service, Docker container, or Tauri desktop app.
- [Privacy Mode](https://docs.neuralcleave.com/privacy-mode.md): One flag routes all LLM traffic through local Ollama, with a persistent, auditable log of everything else that leaves your machine.
- [Changelog](https://docs.neuralcleave.com/changelog.md): Release history and what changed in each version of NeuralCleave.
- [API Overview](https://docs.neuralcleave.com/api-reference/overview.md): NeuralCleave REST API — base URL, authentication, and response format.
- [Chat API](https://docs.neuralcleave.com/api-reference/chat.md): Send messages to NeuralCleave and receive LLM replies.
- [Voice API](https://docs.neuralcleave.com/api-reference/voice.md): REST endpoints for PTT, continuous listening, wake word status, and voice device management.
- [Channels API](https://docs.neuralcleave.com/api-reference/channels.md): List, inspect, and manage connected channel adapters.
- [Settings API](https://docs.neuralcleave.com/api-reference/settings.md): Update LLM provider credentials, active model, thinking level, privacy mode, pipeline, and voice settings at runtime without a gateway restart.
- [MCP API](https://docs.neuralcleave.com/api-reference/mcp.md): Spawn and manage the Model Context Protocol stdio server — exposes NeuralCleave tools to Claude Desktop, Cursor, and other MCP clients.
- [Approvals API](https://docs.neuralcleave.com/api-reference/approvals.md): Review and approve or deny pending shell command executions before they run.
- [Privacy API](https://docs.neuralcleave.com/api-reference/privacy.md): Read and clear the outbound HTTP audit log.

## Optional

- [GitHub](https://github.com/TheAmitChandra/NeuralCleave)
- [Community](https://github.com/TheAmitChandra/NeuralCleave/discussions)