logo
episode-header-image
Feb 2026
51m 37s

MLA 029 OpenClaw

OCDevel
About this episode

OpenClaw is a self-hosted AI agent daemon that executes autonomous tasks through messaging apps like WhatsApp and Telegram using persistent memory. It integrates with Claude Code to enable software development and administrative automation directly from mobile devices.

Links

OpenClaw is a self-hosted AI agent daemon (Node.js, port 18789) that executes autonomous tasks via messaging apps like WhatsApp or Telegram. Developed by Peter Steinberger in November 2025, the project reached 196,000 GitHub stars in three months.

Architecture and Persistent Memory

  • Operational Loop: Gateway receives message, loads SOUL.md (personality), USER.md (user context), and MEMORY.md (persistent history), calls LLM for tool execution, streams response, and logs data.
  • Memory System: Compounds context over months. Users should prompt the agent to remember specific preferences to update MEMORY.md.
  • Heartbeats: Proactive cron-style triggers for automated actions, such as 6:30 AM briefings or inbox triage.
  • Skills: 5,705+ community plugins via ClawHub. The agent can author its own skills by reading API documentation and writing TypeScript scripts.

Claude Code Integration

  • Mobile to Deploy Workflow: The claude-code-skill bridge provides OpenClaw access to Bash, Read, Edit, and Git tools via Telegram.
  • Agent Teams: claude-team manages multiple workers in isolated git worktrees to perform parallel refactors or issue resolution.
  • Interoperability: Use mcporter to share MCP servers between Claude Code and OpenClaw.

Industry Comparisons

  • vs n8n: Use n8n for deterministic, zero-variance pipelines. Use OpenClaw for reasoning and ambiguous natural language tasks.
  • vs Claude Cowork: Cowork is a sandboxed, desktop-only proprietary app. OpenClaw is an open-source, mobile-first, 24/7 daemon with full system access.

Professional Applications

  • Therapy: Voice to SOAP note transcription. PHI requires local Ollama models due to a lack of encryption at rest in OpenClaw.
  • Marketing: claw-ads for multi-platform ad management, Mixpost for scheduling, and SearXNG for search.
  • Finance: Receipt OCR and Google Drive filing. Requires human review to mitigate non-deterministic LLM errors.
  • Real Estate: Proactive transaction deadline monitoring and memory-driven buyer matching.

Security and Operations

  • Hardening: Bind to localhost, set auth tokens, and use Tailscale for remote access. Default settings are unsafe, exposing over 135,000 instances.
  • Injection Defense: Add instructions to SOUL.md to treat external emails and web pages as hostile.
  • Costs: Software is MIT-licensed. API costs are paid per-token or bundled via a Claude subscription key.
  • Onboarding: Run the BOOTSTRAP.md flow immediately after installation to define agent personality before requesting tasks.
Recommended Episodes
Jul 2022
Santiago Valderrama on Getting Smarter on Machine Learning, One Problem at a Time - Ep. 173
Want to learn about AI and machine learning? There are plenty of resources out there to help — blogs, podcasts, YouTube tutorials — perhaps too many. Machine learning engineer Santiago Valdarrama has taken a far more focused approach to helping us all get smarter about the field. ... Show More
27m 15s
Dec 2019
Serverless NLP Model Training
<p class="p1"><span class="s1">Alex Reeves joins us to discuss some of the challenges around building a serverless, scalable, generic machine learning pipeline.<span class= "Apple-converted-space"> </span> The is a technical deep dive on architecting solutions and a discussion of ... Show More
29m 2s
May 2021
474: The Machine Learning House
In this episode, I discuss the architecture of a “machine learning house”, representing the skills and learnings you can use as foundations to build your data science career. Additional materials: www.superdatascience.com/474 
5m 44s
May 2023
Creating instruction tuned models (Practical AI #223)
At the recent ODSC East conference, Daniel got a chance to sit down with Erin Mikail Staples to discuss the process of gathering human feedback and creating an instruction tuned Large Language Models (LLM). They also chatted about the importance of open data and practical tooling ... Show More
26m 33s