xCloudxCloudDownload
Engine (OpenClaw)

Security

Device identity, authentication, sandboxing, and security best practices.

Trust Model

OpenClaw operates under a personal assistant model — one trusted operator per gateway. It is NOT designed for multi-tenant adversarial environments. If you need separate trust boundaries, use separate gateways.

Device Identity

All connections to the engine require ed25519 device signing.

  • Device ID = SHA256 hex of the 32 raw public key bytes
  • Public key = base64url of 32 raw bytes
  • Identity file = ~/.openclaw/identity/device.json

xCloud creates and manages device identity automatically on first launch.

Signing Flow

  1. Server sends a challenge with a nonce
  2. Client signs the nonce with its ed25519 private key
  3. Server verifies the signature and grants access

Authentication Modes

ModeDescription
Token (recommended)Bearer token shared between client and gateway
PasswordVia OPENCLAW_GATEWAY_PASSWORD environment variable
Trusted proxyFor reverse proxies with identity headers
NoneNo auth (testing only, localhost)

xCloud uses token mode by default, generating a random token on first launch.

DM Policies

Control who can message your agents:

PolicyDescription
pairing (default)1-hour pairing code for unknown senders
allowlistOnly approved senders
openAnyone (requires explicit "*")
disabledIgnore all DMs

Sandboxing

Agents can run in Docker containers for isolation:

ModeDescription
allDocker for everything
non-mainDocker only for non-main agents
offNo sandbox, host execution

Workspace Access

LevelDescription
noneNo access to workspace
roRead-only mount
rwRead/write mount

Security Audit

Run a security audit from the command line:

openclaw security audit          # Basic audit
openclaw security audit --deep   # Deep audit
openclaw security audit --fix    # Auto-fix issues

Covers: filesystem permissions, gateway config, hooks, browser, sandbox, tools, and plugins.

Prompt Injection Mitigations

  • DMs blocked by default (pairing/allowlists)
  • Mention gating in groups
  • Links and attachments treated as hostile
  • Sandbox for sensitive execution
  • Secrets stored outside accessible filesystem
  • Use latest-generation models for agents with tools

Credential Storage

ItemLocation
Main config~/.openclaw/openclaw.json
Device identity~/.openclaw/identity/device.json
API keysEnvironment variables or .env file
Secrets~/.openclaw/secrets.json

All directories use permission 700, files use 600.