CraftOS-dev/CraftBot

One agent. Every kind of work.

What it solves

CraftBot is a general-purpose AI agent harness designed to move beyond simple chat and tool calls. It solves the problem of static AI assistants by allowing an agent to build, evolve, and operate its own custom SaaS-like tools (Agent Apps) to automate complex workflows, manage data, and proactively assist users with their daily goals.

How it works

CraftBot functions as a background service that integrates with various LLM providers (OpenAI, Gemini, Claude, Ollama, etc.) and connects to external applications via OAuth or API keys. It uses a hybrid memory system combining RAG, a knowledge graph, and an Agent File System to remember user preferences and consolidate events. A key feature is the "Agent App" system, where the agent can scaffold a data model, backend API, and React UI from plain language descriptions, or import existing codebases, creating a living UI that the agent can read and manipulate directly.

Who it’s for

It is designed for users who want a proactive AI employee capable of managing tasks across multiple platforms, as well as developers and power users who want to create custom, AI-driven dashboards and automation tools without relying on solely on third-party SaaS subscriptions.

Highlights

  • Agent Apps: Ability to build, import, or evolve custom internal apps (like Kanban boards or CRMs) that the agent can operate.
  • Proactive Assistance: Learns habits and goals to initiate planning and tasks autonomously.
  • Extensive Ecosystem: Includes 40+ pre-defined agent profiles, 120+ playbooks, 150+ MCPs, and 170+ skills.
  • Hybrid Memory: A "second brain" that uses RAG and knowledge graphs to maintain long-term context.
  • Flexible Deployment: Supports self-hosting via BYOK or local models through Ollama, with both browser and CLI interfaces.

Related

  • Project
    zhayujie/CowAgent

    CowAgent is an open‑source, self‑hosted AI assistant that plans tasks, runs tools, keeps a personal knowledge base, and can operate as a team of cooperating agents. It supports many LLM providers, multimodal input, dozens of chat channels (Telegram, Slack, WeChat, etc.), a plug‑in skill system, and a three‑tier memory with nightly “Deep Dream” distillation. Install via a one‑line script or Docker, manage everything through a web console, and extend the system through the public Skill Hub or custom tool integrations.

  • Project
    DotHarness/dotcraft

    DotCraft is an open‑source AI‑agent platform you run locally. It provides a desktop app, CLI, Android client, and SDKs (TypeScript/.NET) so the same agent can plan, execute, and coordinate sub‑agents across devices, integrate with IDEs, chat platforms, and embed into your own applications.

  • Project
    ruvnet/metaharness

    MetaHarness is a CLI and web studio that creates a repo‑specific AI‑agent harness (custom CLI, skills, memory, and policy) which can run on multiple host agents and be published as an npm package.

  • Project
    NousResearch/hermes-agent

    Hermes Agent is a self-improving AI agent from Nous Research that builds persistent memory, creates and refines skills from experience, and runs across CLI and messaging platforms. It solves the problem of assistants that forget context and don't improve, and can run on servers or serverless infrastructure.

  • Project
    siddsachar/row-bot

    Row‑Bot is a self‑hosted desktop AI assistant that lets you converse with LLMs to create files, run code, manage designs, and interact with services—all while keeping data locally. It offers memory via a personal knowledge graph, goal‑driven agents, approval‑gated actions, multi‑device remote access, and integrations (GitHub, Notion, Slack, etc.). Runs on Windows/macOS/Linux or Docker, using Ollama or any OpenAI‑compatible provider, and stores everything securely on your machine.