Command-line tools
The two axonpush CLIs, the wizard launcher that installs the integration skills and opens your coding agent, and the self-host installer that provisions the whole stack into your AWS account.
axonpush ships two command-line entry points. Neither is a client library, the SDKs cover that; these are for wiring a project up and for running your own deployment.
The wizard launcher
npx @axonpush/wizardA thin shim that installs the integration skills and opens the first
coding agent it finds (claude, cursor or codex) so you can run
/axonpush-integrate. It exists because the one-liner is on the README and the
marketing site, and it does nothing the skills bundle does not.
--local <path> installs from a local checkout instead of GitHub, for working on
the skills themselves.
See Let a coding agent wire it up for what the integrate skill does.
The self-host installer
curl -fsSL https://get.axonpush.xyz/install.sh | shProvisions the whole axonpush stack into your own AWS account, the Go Lambdas,
the Bun authsvc, Aurora DSQL, S3, SQS and CloudFront, and self-bootstraps the
database. It prompts for your licence key (or reads LICENSE_KEY), downloads the
signed release bundle of prebuilt artifacts, and prints your instance URL. Needs
only the AWS CLI, Node and Go, no Docker, no container registry, no secrets to
pre-create. Re-run it to update in place.
The full walkthrough, including the custom-domain and teardown steps, is on the Self-host on your own AWS page.
Related
Self-host on your own AWS
Run the full axonpush stack in your own AWS account, serverless, pay-per-use, and roughly $0/month at idle. One licence key, one command, your data never leaves your cloud.
Add observability to an agent framework
Drop-in handlers for LangChain, OpenAI Agents, Anthropic, CrewAI, Deep Agents, LangGraph, LlamaIndex, Mastra, Vercel AI SDK, Google ADK and Semantic Kernel, plus what each one actually emits.