AlphaClaw is a management harness around OpenClaw that replaces hand-edited config files and SSH sessions with a password-protected web dashboard. It runs OpenClaw's gateway as a managed child process and layers operations tooling on top: a guided onboarding wizard (model selection, provider credentials, GitHub repo, channel pairing), multi-agent management from a sidebar UI, and one-click deploy templates for Railway and Render. It currently targets Docker/Linux deployments, and the project stresses there is no lock-in — remove AlphaClaw and the underlying OpenClaw agent keeps running unchanged.
Reliability and auditability are the core pitch. A watchdog performs periodic health checks, detects crashes and crash loops, auto-repairs by running openclaw doctor and relaunching the gateway, and sends alerts to Telegram, Discord, or Slack, with a SQLite-backed incident log and a live browser terminal. Git sync commits the OpenClaw workspace to GitHub on a schedule, and bundled anti-drift prompts (AGENTS.md, TOOLS.md) are injected into the agent's system prompt to enforce commit discipline and change summaries — together making agent actions version-controlled and reviewable. The dashboard also covers usage/cost breakdowns, cron job management with run history, a workspace file explorer with diff view, webhook endpoints with transform modules, guided Google Workspace OAuth (including Gmail watch via Pub/Sub), and local-node setup for VPS deployments. An optional OpenAI-compatible /v1 proxy can expose the gateway to trusted server-to-server callers.
The README is candid that AlphaClaw trades some of OpenClaw's default hardening for convenience: a single setup password gates all access, channel pairings can be approved from the UI, the first CLI pairing is auto-approved, and webhook tokens may be passed via query string. Users who want OpenClaw's full security posture are advised to run it without the wrapper.
