OpenClaw is an open-source personal AI agent created by Peter Steinberger that runs locally on your own machine (macOS, Linux, Windows) rather than in a vendor’s cloud. It’s pitched as “a 24/7 assistant with access to its own computer” — the emphasis is on doing real things (sending emails, filling forms, running scripts), not just chatting.
Core design choices
| Choice | What it buys you |
|---|---|
| Runs locally | State (memory, credentials, task history) stays on your machine, not a SaaS vendor’s servers |
| Model-agnostic gateway | Swap Claude, GPT, or a local model behind the same agent without rewriting anything |
| Multi-channel | One agent, reachable from 29+ platforms — WhatsApp, Telegram, Discord, Slack, Signal, iMessage |
| Skills framework | Community-contributed or self-authored plugins via ClawHub; the agent can write its own skills |
You (WhatsApp / Slack / Discord / ...)
│
▼
OpenClaw gateway ──▶ Model of your choice (Claude / GPT / local)
│
├── Browser control (fill forms, extract data)
├── Filesystem read/write
├── Shell / script execution
└── Cron jobs (proactive, unattended tasks)
What people actually use it for
Reported use cases skew toward personal ops automation: inbox triage, calendar coordination, flight check-ins, expense reporting and tax prep, website building, code review, test automation, and orchestrating multiple agents across a Discord server.
The positioning: self-hosted vs. SaaS agent
OpenClaw’s pitch is explicitly a reaction against cloud agent platforms: “your context and skills live on YOUR computer, not a walled garden.” That’s the same self-hosting argument What is Hermes Agent makes, and both projects compete on the same axis — control and data locality — rather than raw capability.
| Axis | OpenClaw | What is Hermes Agent |
|---|---|---|
| Origin | Peter Steinberger (indie) | Nous Research (lab, tied to Hermes models) |
| Memory model | Persistent, skills-framework driven | ”Holographic” fact memory (SQLite/FTS5) with a closed-loop skill-distillation step |
| Distribution channels | 29+ chat platforms | Telegram, Discord, Slack, WhatsApp, Signal, CLI |
| Extension model | ClawHub community skills/plugins | Auto-distilled skills from its own task history |
Note on claims: OpenClaw’s own marketing cites very large adoption numbers (GitHub stars in the hundreds of thousands) for a project this young. Treat vendor-reported growth figures for any fast-moving open-source AI project as directional, not verified, until you’ve checked the repo yourself.
Why it’s on the backlog: it’s the most “consumer-facing” entry in this list — useful as a reference point for how far a self-hosted agent can reach into personal systems (browser + shell
- filesystem) before you even get to enterprise frameworks like What is Google ADK or What is CrewAI.
Local graph
Linked from 3 notes
What is Harness Engineering
The discipline of designing everything that wraps a raw LLM into a reliable agent — tool contracts, system prompt architecture, context/memory management, permission gates, and feedback loops. Distinct from prompt engineering and model training.
What is Hermes Agent
Nous Research's open-source, self-hosted AI agent that runs continuously on your own server, builds persistent memory over time, and distills its own reusable 'skills' from completed tasks.
Agentic AI: Projects & Engineering Mastery
A book-shaped table of contents for Agentic AI: Projects & Engineering Mastery: hands-on practitioner builds, Principal/Staff-level technical leadership, and the lookup appendices and vendor/framework reference notes for the whole series. Book 6 of the AI Systems Engineering series.
Related notes
What is Google ADK
Google's open-source, code-first Agent Development Kit — a multi-language framework for building, evaluating, and deploying agents, positioned as an 'agent execution framework' rather than a toolkit.
What is CrewAI
Open-source Python framework for multi-agent orchestration — role-based 'Crews' for autonomous collaboration and event-driven 'Flows' for precise control, now a de facto standard for production agentic pipelines.
What is Hermes Agent
Nous Research's open-source, self-hosted AI agent that runs continuously on your own server, builds persistent memory over time, and distills its own reusable 'skills' from completed tasks.
What is Gemini Enterprise Agent Platform
Google's April 2026 unification of agent tooling — a four-stage lifecycle (build, scale, govern, optimize) wrapping Agent Studio/ADK, a stateful Agent Runtime, an Identity/Registry/Gateway governance stack, and native A2A + MCP interop.