qaml-ai/camelAI

camelAI — an AI coding assistant platform built on Cloudflare Workers and Durable Objects

What it solves

camelAI provides teams with persistent AI coding workspaces that go beyond simple chat. It solves the problem of ephemeral AI interactions by providing a persistent environment where coding agents have access to project files, connected external data, and the ability to build and deploy applications directly from the workspace.

How it works

The platform uses a custom agent harness built on Cloudflare's infrastructure. Each chat thread runs its own coding agent within a Durable Object. The agent interacts with a workspace filesystem (stored in SQLite and R2) and executes JavaScript tools in isolated V8 isolates. For heavier tasks like application builds, notebook analysis, and SQL execution, the system spins up short-lived Cloudflare sandbox containers. Completed projects can be published to managed app hosts via Workers for Platforms.

Who it’s for

It is designed for development teams who need an integrated AI coding environment that can manage project state, connect to various APIs and databases, and deploy live applications.

Highlights

  • Persistent Workspaces: Chat threads and project files are maintained across sessions.
  • Integrated Deployment: Ability to build and publish applications to production through a managed dispatcher.
  • Extensible Connectivity: Built-in integrations for Slack, Discord, email, databases, and various APIs.
  • Flexible Model Support: Compatible with providers including Anthropic, OpenAI, OpenRouter, and Bedrock.
  • Isolated Execution: Uses a combination of V8 isolates and sandbox containers for secure tool execution.

Related

  • Project
    Team-Commonly/commonly

    An open-source workspace for collaborating with AI agents from any runtime, providing them with persistent identity and shared project memory to eliminate redundant context-setting.

  • Project
    chadbyte/clay

    A self-hosted workspace that turns local coding agents into a shared, persistent team, allowing people and AI agents to collaborate across different providers in one browser-based environment.

  • Project
    aerovato/container

    A tool that provides persistent, isolated Docker or Podman workspaces for AI coding agents to prevent system pollution and limit destructive operations.

  • Project
    rcarmo/piclaw

    A self-hosted AI workspace that integrates the Pi Coding Agent with a persistent web UI, built-in development tools, and multi-provider LLM support.

  • Project
    openagents-org/openagents

    An open-source collaborative OS for AI agents that provides a unified workspace where multiple agents from different runtimes can share context, files, and a browser.