# OpenPawz

> Your AI, your rules. The open-source AI agent OS with 25,000+ integrations via MCP bridge, 12 channels, multi-agent orchestration — all running locally on your desktop.

## Docs

- [Discord](https://docs.openpawz.ai/channels/discord.md): Connect a Discord bot to your Pawz agents.
- [Discourse](https://docs.openpawz.ai/channels/discourse.md): Connect your agent to a Discourse community forum.
- [IRC](https://docs.openpawz.ai/channels/irc.md): Connect to IRC networks so users can chat with your agents in IRC channels.
- [Matrix](https://docs.openpawz.ai/channels/matrix.md): Connect to a Matrix homeserver so users can chat with your agents via any Matrix client (Element, FluffyChat, etc.).
- [Mattermost](https://docs.openpawz.ai/channels/mattermost.md): Connect to a self-hosted Mattermost instance.
- [Nextcloud](https://docs.openpawz.ai/channels/nextcloud.md): Connect to Nextcloud Talk so users on your Nextcloud instance can chat with Pawz agents.
- [Nostr](https://docs.openpawz.ai/channels/nostr.md): Connect to the Nostr decentralized network so users can chat with your agents via any Nostr client.
- [Channels](https://docs.openpawz.ai/channels/overview.md): Connect your OpenPawz agents to 12 messaging platforms — all running simultaneously from one app.
- [Slack](https://docs.openpawz.ai/channels/slack.md): Connect a Slack bot to your Pawz agents.
- [Telegram](https://docs.openpawz.ai/channels/telegram.md): Connect a Telegram bot to your Pawz agents.
- [Twitch](https://docs.openpawz.ai/channels/twitch.md): Connect to Twitch chat so viewers can interact with your agents during streams.
- [Webchat](https://docs.openpawz.ai/channels/webchat.md): Webchat is a built-in HTTP/WebSocket chat interface you can embed on your own website or access directly.
- [WhatsApp](https://docs.openpawz.ai/channels/whatsapp.md): Give your agent a WhatsApp number. People message that number, your agent replies — all on your machine.
- [Agents](https://docs.openpawz.ai/guides/agents.md): Agents are the core of Pawz — autonomous AI personas that chat, use tools, remember context, and collaborate.
- [Automations](https://docs.openpawz.ai/guides/automations.md): Automations are recurring tasks that run on a schedule using cron expressions.
- [Browser Sandbox](https://docs.openpawz.ai/guides/browser.md): The browser sandbox lets agents browse the web in a controlled, auditable environment.
- [Container Sandbox](https://docs.openpawz.ai/guides/container-sandbox.md)
- [Content Studio](https://docs.openpawz.ai/guides/content-studio.md)
- [Dashboard & Today](https://docs.openpawz.ai/guides/dashboard.md)
- [Email](https://docs.openpawz.ai/guides/email.md): Pawz can read and send email through your existing accounts using Himalaya, an IMAP/SMTP CLI client.
- [Extensions](https://docs.openpawz.ai/guides/extensions.md)
- [Flows](https://docs.openpawz.ai/guides/flows.md): Build visual AI workflows by connecting integrations on a drag-and-drop canvas — triggers, agents, tools, conditions, code, and outputs.
- [Foundry — Models & Chat Modes](https://docs.openpawz.ai/guides/foundry.md): The Foundry is your control center for managing AI providers and creating reusable chat mode presets. It provides two tabs: Models and Chat Modes.
- [Google Workspace](https://docs.openpawz.ai/guides/google-workspace.md): Connect Pawz to Gmail, Calendar, Drive, Sheets, and Docs — just click Connect with Google.
- [Integration Guardrails](https://docs.openpawz.ai/guides/integration-guardrails.md): Safety controls for agent-driven integrations: action risk classification, per-service rate limits, agent permissions, dry-run plans, and credential audit logging.
- [Integration Health](https://docs.openpawz.ai/guides/integration-health.md): Monitor connection status, token expiry, failure rates, and get smart suggestions for connected integrations.
- [Integrations](https://docs.openpawz.ai/guides/integrations.md)
- [MCP Bridge](https://docs.openpawz.ai/guides/mcp-bridge.md): Zero-Gap Automation — how OpenPawz connects to 25,000+ integrations via an embedded n8n engine and the Model Context Protocol.
- [Memory](https://docs.openpawz.ai/guides/memory.md): Project Engram — a biologically-inspired three-tier memory architecture with hybrid search, PII encryption, and graph-based knowledge management.
- [Modular Workspace](https://docs.openpawz.ai/guides/modular-workspace.md)
- [Orchestrator](https://docs.openpawz.ai/guides/orchestrator.md): The Orchestrator enables multi-agent projects where a boss agent delegates tasks to worker agents.
- [PawzHub — Marketplace](https://docs.openpawz.ai/guides/pawzhub.md)
- [Model Pricing & Budget](https://docs.openpawz.ai/guides/pricing.md): Pawz tracks token usage and costs across all AI providers, enforces daily budgets, and can automatically route tasks to cheaper models when appropriate.
- [Projects — File Browser](https://docs.openpawz.ai/guides/projects.md): The Projects view is a local file browser built into Pawz. It lets you browse project folders, inspect Git status, and preview text files — all within the app using Tauri's filesystem APIs.
- [Research](https://docs.openpawz.ai/guides/research.md): The Research view lets agents conduct structured web research with source tracking and credibility scoring.
- [Session Management](https://docs.openpawz.ai/guides/sessions.md)
- [Skills](https://docs.openpawz.ai/guides/skills.md): Skills are the foundation of the Pawz extensibility system — part of the 400+ native integrations that ship built-in, extended to 25,000+ via the MCP Bridge.
- [Slash Commands](https://docs.openpawz.ai/guides/slash-commands.md): Type `/` in any chat to access quick commands. Commands are organized by category.
- [Tailscale](https://docs.openpawz.ai/guides/tailscale.md): Tailscale integration lets you access Pawz remotely and expose agent services over your tailnet.
- [Tasks](https://docs.openpawz.ai/guides/tasks.md): The Tasks view is a kanban board for managing work items that agents can pick up and execute.
- [Trading](https://docs.openpawz.ai/guides/trading.md): Pawz includes a trading dashboard for cryptocurrency operations via Coinbase, Ethereum DEX (Uniswap V3), and Solana DEX (Jupiter + PumpPortal).
- [Voice & TTS](https://docs.openpawz.ai/guides/voice.md): Pawz supports text-to-speech so agents can speak their responses aloud.
- [Anthropic](https://docs.openpawz.ai/providers/anthropic.md): Use Claude Opus 4, Claude Sonnet 4, and other Anthropic models.
- [Custom Provider](https://docs.openpawz.ai/providers/custom.md): Connect any OpenAI-compatible API as a custom provider.
- [DeepSeek](https://docs.openpawz.ai/providers/deepseek.md): Use DeepSeek's reasoning and coding models.
- [Google Gemini](https://docs.openpawz.ai/providers/google.md): Use Gemini 2.5 Pro, Gemini 2.5 Flash, and other Google models.
- [Grok](https://docs.openpawz.ai/providers/grok.md): Use xAI's Grok models.
- [Mistral](https://docs.openpawz.ai/providers/mistral.md): Use Mistral AI's models.
- [Moonshot](https://docs.openpawz.ai/providers/moonshot.md): Use Moonshot AI's Kimi models.
- [Ollama](https://docs.openpawz.ai/providers/ollama.md): That's it — no API key or configuration needed.
- [OpenAI](https://docs.openpawz.ai/providers/openai.md): Use GPT-4.1, o3, o4-mini, and other OpenAI models.
- [OpenRouter](https://docs.openpawz.ai/providers/openrouter.md): OpenRouter supports hundreds of models. Some popular ones:
- [Providers](https://docs.openpawz.ai/providers/overview.md): 10 AI providers and any OpenAI-compatible API — local or cloud, cheap or powerful.
- [The Agent Canvas](https://docs.openpawz.ai/reference/agent-canvas.md): Generative UI, Live Observability, and Agent-Built Dashboards — an architecture for agents that don't just talk, they show.
- [Architecture](https://docs.openpawz.ai/reference/architecture.md): OpenPawz is a Tauri v2 desktop application with a Rust backend and TypeScript frontend.
- [The Foreman Protocol](https://docs.openpawz.ai/reference/foreman-protocol.md): Low-Cost Tool Execution via Any Worker Model and Self-Describing MCP — a novel architecture for reducing tool execution costs by delegating to cheaper models.
- [The Librarian Method](https://docs.openpawz.ai/reference/librarian-method.md): Intent-Stated Tool Discovery via Embeddings — a novel approach to solving tool bloat in agentic AI systems.
- [Quality & Testing](https://docs.openpawz.ai/reference/quality.md): OpenPawz maintains enterprise-grade quality gates. Every commit is validated by a 3-job CI pipeline and must pass all checks before merging.
- [Security](https://docs.openpawz.ai/reference/security.md): OpenPawz is designed with defense-in-depth: 20+ security layers protect against prompt injection, data exfiltration, SSRF, MITM attacks, memory forensics, agent fixation, inter-agent manipulation, and unauthorized actions.
- [Troubleshooting](https://docs.openpawz.ai/reference/troubleshooting.md): Common issues and how to fix them.
- [Configuration](https://docs.openpawz.ai/start/configuration.md): Optional configuration for providers, workspace, environment, and agent defaults.
- [Dashboard](https://docs.openpawz.ai/start/dashboard.md): The Today view — your AI agent command center at a glance.
- [Features](https://docs.openpawz.ai/start/features.md): The most comprehensive AI agent platform ever built — 25,000+ integrations via MCP bridge, multi-agent, multi-channel, voice, trading, memory, and security.
- [Create Your First Agent](https://docs.openpawz.ai/start/first-agent.md): Agents are the core of OpenPawz. Each agent has its own personality, AI model, tools, and conversation history.
- [Connect Your First Provider](https://docs.openpawz.ai/start/first-provider.md): OpenPawz needs at least one AI provider to power your agents. Use a local model (Ollama — free) or a cloud API.
- [OpenPawz](https://docs.openpawz.ai/start/getting-started.md): Your AI, your rules. The open-source AI agent OS with 25,000+ integrations via MCP bridge — private, powerful, extensible.
- [Installation](https://docs.openpawz.ai/start/installation.md): Get OpenPawz running on macOS, Linux, or Windows in under 5 minutes.
- [Overview](https://docs.openpawz.ai/start/overview.md): The open-source AI agent OS — 25,000+ integrations via MCP bridge, multi-agent, multi-channel, fully local, infinitely extensible.
- [Showcase](https://docs.openpawz.ai/start/showcase.md): What you can build with OpenPawz — use cases, feature highlights, and real workflows.

## OpenAPI Specs

- [openapi](https://docs.openpawz.ai/api-reference/openapi.json)