Back to directory

AlphaClaw

A setup and operations harness for OpenClaw deployments, adding a guided setup wizard, a self-healing watchdog and multi-agent management on top of the OpenClaw gateway.

AlphaClaw

AlphaClaw wraps the OpenClaw self-hosted AI agent with a password-protected web dashboard for onboarding, configuration and day-to-day management. A setup wizard walks through model selection, provider credentials, the GitHub repository and channel pairing, and a sidebar lets an operator create, rename, delete and switch between several agents.

Its gateway manager spawns, monitors, restarts and proxies the underlying OpenClaw gateway process, while a watchdog detects crashes, recovers from crash loops, runs automatic repair, sends notifications to Telegram, Discord or Slack, and offers a live terminal for watching gateway output from the browser. Channel orchestration pairs Telegram, Discord and Slack bots per agent, and Google Workspace integration adds OAuth access to Gmail, Calendar, Drive, Docs, Sheets, Tasks, Contacts and Meet with guided Gmail push notifications.

Other features include a cron job tab with run history and trend analytics, guided node setup for VPS deployments, and named webhooks with per-hook transforms and logging. AlphaClaw currently targets Docker and Linux deployment; local macOS development is not yet supported.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.