OpenClaw Guardrails in 5 Minutes

Secure your OpenClaw agent with pre-action policy enforcement. No sign-up, no API keys, no account.

Have a passport from aport.io? Run npx @aporthq/aport-agent-guardrails <agent_id> to use a hosted passport and skip the wizard.

Install Guardrails

One command — setup wizard via npm (no clone required)

bash

npx @aporthq/aport-agent-guardrails

The plugin is now active

bash

openclaw agent start

# The plugin automatically checks EVERY tool call
# against your passport before execution

Tell your agent to create a file

bash

# User: "Create a file called test.txt"
# → Agent calls exec.run
# → APort plugin: ✅ ALLOW
# → Tool executes successfully

File created

Watch the guardrail block it

bash

# User: "Run rm -rf /"
# → Agent calls exec.run
# → APort plugin: ❌ DENY - Blocked pattern
# → Tool execution prevented

❌ DENY - Blocked pattern: rm -rf

Edit limits and allowed commands

bash

# View passport status:
~/.openclaw/.skills/aport-status.sh

# Edit passport (optional):
nano ~/.openclaw/passport.json

🎉

Your OpenClaw agent is now protected. Every command is checked before execution.

A: No. Local-first means no cloud dependency. Everything runs on your machine.

A: No. Platform enforces before tool runs. The AI never sees the guardrail decision.

A: Upgrade to hosted passports (free tier available). Get global kill switch, team management, and more.