Aller au contenu
TakoAPI
Retour aux skills

moltbot-security

par nextfrontierbuilds

Moltbook
2.3kTéléchargements rapportés par clawskills.sh
3Étoiles
0Vues
0 mention J'aimeConnecte-toi pour aimer ce skill
—Pas encore noté
Ta note

Résumé

Security hardening guide for AI agents.

Description

Ce que fait ce skill

A configuration and hardening guide for OpenClaw and Moltbot AI agent gateways. Addresses publicly exposed gateways, which can leak API keys, conversation histories, and shell access to any internet scanner. Based on research that found 1,673+ exposed instances on Shodan.

It consolidates the most exploited gateway misconfigurations into a single audit command and config template, replacing ad-hoc research when deploying AI agents on cloud servers.

Quand l'utiliser

  • Locking down an OpenClaw gateway before exposing it beyond localhost
  • Auditing an existing Moltbot setup for common misconfigurations
  • Setting up Tailscale to access your AI agent remotely without opening a public port
  • Hardening a cloud server running OpenClaw with UFW firewall rules
  • Preventing prompt injection attacks that exfiltrate private conversation histories

Exemple de workflow

ENTRÉE

User asks: audit my OpenClaw gateway and fix any security issues

AGENT
  1. 1

    Runs openclaw security audit --deep to detect binding, auth, and permission issues

  2. 2

    Identifies gateway bound to all interfaces instead of loopback

  3. 3

    Applies openclaw security audit --deep --fix to correct file permissions and binding

  4. 4

    Generates a secure token with openssl rand -hex 32 and writes it to the config

  5. 5

    Verifies the updated config matches the secure defaults template

SORTIE

Gateway rebound to loopback, auth token set, file permissions corrected to 600/700, mDNS broadcasting disabled

Prérequis

  • Tailscale account (only required if using remote access via Tailscale)

Installation

clawhub install nextfrontierbuilds/moltbot-security

Exécute la commande CLI, ou colle l'instruction dans Claude Code, Cursor, Codex ou l'agent de code que tu utilises.

Liens