OpenClaw Ecosystem: Categories, What to Use, and What to Avoid
The ecosystem is growing fast, which is good for innovation and bad for supply-chain risk.
This guide helps you understand the categories and choose tools that match your risk tolerance.
Where to browse the ecosystem
Start here: /ecosystem
The main categories (how to interpret them)
- Foundation: core registries, L2/platform primitives, identity/auth rails.
- Infrastructure: messaging, storage, developer platforms, protocols.
- Tools: dev tools, monitoring, helper apps, orchestration utilities.
- Security & Trust: verification, auditing, threat intel, incident tooling.
- Compute & Data: model runtimes, indexes, data planes.
- Simulation & Agent Wars: sandboxes, arenas, multi-agent games.
- Social & Community: networks and community surfaces.
- DeFi & Crypto: tokens, markets, onchain escrow, agent trading.
How to choose a project (fast filter)
- Does it require wallet signing or token transfers? Treat as high risk.
- Does it require broad permissions (network + shell)? Sandbox it.
- Does it have clear docs, ownership, and intent?
- Can you use it read-only first?
The safety baseline
- verify skills: /verifier
- prefer audited: /verified-skills
- sandbox: /guides/sandbox-setup