camelAI
camelAI — an AI coding assistant platform built on Cloudflare Workers and Durable Objects
What is it?
What it is
An AI coding assistant providing persistent workspaces, connected data, and deployable applications.
Why it exists
To provide teams with persistent AI coding workspaces backed by Cloudflare, featuring workspace-aware agents, external service integrations, and managed application hosting.
Who should use it
Software development teamsAI engineersDevOps engineersPlatform engineersOrganizations requiring persistent AI coding workspaces
Who should avoid it
Users without a Cloudflare account (for cloud deployment)Users unable to run Docker (for sandbox-backed features and agent evals)Developers looking for a simple CLI-only tool (this is a full platform)
How it works
A quick walkthrough in plain English
How camelAI works
Step 1 of 3
You interact with it
Open camelAI, send a request, or connect it to your stack.
Features
Advantages
- Persistent, isolated workspaces for team collaboration
- Scalable infrastructure via Cloudflare's global network
- Seamless integrations with third-party services
- Flexible deployment options (cloud or self-hosted)
- Security-focused design (credentials outside execution sandbox)
- Modular architecture for customization
- Support for multiple AI model providers
- Self-hosting for private networks and on-premises use
Disadvantages
- Requires Cloudflare account and specific infrastructure (Docker, Node.js 22)
- Complex setup and maintenance for self-hosted instances
- Limited outbound email functionality in self-hosted mode
- Dependency on specific tools (Bun, Vite) may limit accessibility
- No multi-node failover in self-hosted deployments
- Potential learning curve for advanced features and architecture
Installation
cloud
bun run deploy:main:staging bun run deploy:main:prod bun run deploy:dispatcher:staging bun run deploy:dispatcher:prod
docker
bun run selfhost:init bun run selfhost:doctor bun run selfhost:up
native
git clone https://github.com/qaml-ai/camelAI.git cd camelAI bun install --frozen-lockfile cp.dev.vars.example.dev.vars # Replace the placeholder signing and encryption secrets in.dev.vars. # For hosted model access, also add a development Cloudflare AI Gateway token. # Start the app with local authentication: bun run dev:local-auth # Or to exercise the normal OAuth flow: bun run dev
compose
bun run selfhost:init bun run selfhost:doctor bun run selfhost:up
FAQ
What are the prerequisites for running camelAI locally?
You need Node.js 22 or newer, Bun, a Cloudflare account with access to development resources, and Docker for sandbox-backed features and agent evals.
How do I start camelAI with a seeded local development user?
Run `bun run dev:local-auth` after installing dependencies and configuring `.dev.vars`; this starts the Vite dev server and seeds a `Local Dev` user, organization, and workspace on http://localhost:3001.
Which command runs the full test suite including application, shared-library, and Worker tests?
Use `bun run test:all` to run both the application/shared-library test suite and the Worker/Durable Object test suite.
How can I self-host camelAI for private networks or on-premises evaluation?
Execute `bun run selfhost:init`, then `bun run selfhost:doctor`, and finally `bun run selfhost:up`; see SELF_HOSTING.md and infra/selfhost/README.md for detailed configuration.
What deployment commands are available for the main application and the dispatcher?
Main application: `bun run deploy:main:staging` and `bun run deploy:main:prod`. Dispatcher: `bun run deploy:dispatcher:staging` and `bun run deploy:dispatcher:prod`. Other Workers have dedicated `deploy:*` scripts in package.json.
Featured in Videos
YouTube tutorials and walkthroughs for camelAI
Alternatives
Similar projects ranked by category, topics, and text overlap.
