Skip to content

QUICKSTART.md

Latest commit

 

History

History
185 lines (132 loc) · 4.01 KB

QUICKSTART.md

File metadata and controls

185 lines (132 loc) · 4.01 KB
Raw

AnyClaw Quickstart

This guide covers a source checkout of AnyClaw and focuses on commands that are available in this repository.

Prerequisites

  • Go 1.25 or newer
  • Node.js 22 or newer, with Corepack enabled, when building or running the control UI from source
  • Docker, when using the Compose-based gateway runtime

Build the CLI

From the repository root:

go build -o anyclaw ./cmd/anyclaw

On Windows, use an .exe suffix if you want a directly runnable binary:

go build -o anyclaw.exe ./cmd/anyclaw

Then verify the command list:

./anyclaw --help

On Windows:

.\anyclaw.exe --help

Configure a Provider

Run the first-run setup wizard:

./anyclaw onboard

For non-interactive environments, write defaults without prompts:

./anyclaw onboard --non-interactive

Check the resulting configuration:

./anyclaw doctor
./anyclaw models status
./anyclaw config validate

By default, onboard writes anyclaw.json in the current directory, and commands read that file unless you pass --config <path>. The committed anyclaw.example.json is only a template/reference file.

Environment Variables

.env.example is a template for values you export in your shell. It also includes the ANYCLAW_LLM_* names that docker-compose.yml reads from a Compose-managed .env file. The AnyClaw CLI does not load .env files by itself.

For local CLI runs, export variables before launching AnyClaw. Common examples:

export LLM_PROVIDER=openai
export LLM_MODEL=gpt-4o
export OPENAI_API_KEY=...

For Docker Compose, the provided compose file reads ANYCLAW_LLM_PROVIDER, ANYCLAW_LLM_MODEL, and ANYCLAW_LLM_API_KEY from the Compose environment or a Compose-managed .env file.

Build or Run the Control UI

The gateway serves the control UI. When running from a repository checkout, the gateway can auto-build the UI if Node.js is available. You can also build it explicitly:

corepack enable
corepack prepare pnpm@10.23.0 --activate
pnpm run ui:build

Useful UI scripts:

pnpm run ui:dev
pnpm run ui:test
pnpm run ui:preview

Launch the Desktop Shell on Windows

Double-click AnyClaw Desktop.cmd from the repository root.

The launcher starts cmd/anyclaw-desktop/build/bin/anyclaw-desktop.exe when it already exists. If the executable is missing, it runs scripts/build-desktop.ps1 first, then starts the desktop shell. Build and launch output is written to .anyclaw/logs/desktop-launch.log.

Command-line equivalents:

pnpm run desktop:start
pnpm run desktop:build
powershell -NoProfile -ExecutionPolicy Bypass -File scripts/start-desktop.ps1 -Rebuild

Run the Gateway Locally

Start the gateway in the foreground:

./anyclaw gateway run --host 127.0.0.1 --port 18789

gateway start is accepted as an alias for the same foreground command:

./anyclaw gateway start --host 127.0.0.1 --port 18789

Run it as a daemon:

./anyclaw gateway daemon start
./anyclaw gateway daemon stop

Inspect the running gateway:

./anyclaw gateway status
./anyclaw gateway sessions
./anyclaw gateway events

Run with Docker Compose

Create a .env file for Compose. Use the ANYCLAW_LLM_* names because these are the variables read by docker-compose.yml:

ANYCLAW_LLM_PROVIDER=openai
ANYCLAW_LLM_MODEL=gpt-4o
ANYCLAW_LLM_API_KEY=...

Create the host-side runtime workspace, then start the gateway:

mkdir -p workspace/workflows
docker compose up --build

The service exposes the gateway on port 18789. Gateway state is stored in the anyclaw-data volume at /workspace/.anyclaw, while agent workflow files are bind-mounted from ./workspace/workflows to /workspace/workflows. This runtime workspace is separate from the source checkout used to build the image.

Useful Commands

./anyclaw models list
./anyclaw models set gpt-4o-mini
./anyclaw config file
./anyclaw status
./anyclaw health
./anyclaw sessions
./anyclaw skill list
./anyclaw skill catalog