feelgoodbot for Openclaw

A security-focused tool providing file integrity monitoring and TOTP step-up authentication for macOS-based AI agents.

kris-hansen
v1.1.0
Feb 9, 2026
1
2.5k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install feelgoodbot

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install feelgoodbot using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is feelgoodbot?

feelgoodbot is a comprehensive security utility designed specifically for the macOS ecosystem. It serves two primary functions: a robust File Integrity Monitor (FIM) and a Time-based One-Time Password (TOTP) step-up authentication system. By integrating this into Openclaw Skills, users can ensure their system files remain untampered while adding a critical human-in-the-loop verification layer for sensitive AI agent actions.

This tool is essential for developers using AI agents who want to prevent unauthorized system modifications or malware persistence. It allows agents to perform high-stakes operations—like sending emails or making payments—only after the user provides a verified OTP, effectively bridging the gap between agentic autonomy and user security.

feelgoodbot Use Cases

  • Monitoring critical system binaries and shell configurations for unauthorized tampering.
  • Requiring manual OTP approval for AI-driven financial transactions or payment processing.
  • Protecting sensitive file deletions or SSH access through a 15-minute cached security session.
  • Notifying users via webhooks when persistence mechanisms like launch daemons are modified.

How feelgoodbot Works

  1. The user initializes a baseline snapshot of the system's state using the CLI to establish a known secure configuration.
  2. A background daemon continuously monitors specified directories and configuration files for any unauthorized changes.
  3. Before performing a sensitive operation, the AI agent checks the step-up requirement for the specific action pattern.
  4. If authentication is required, the agent prompts the user for a 6-digit code, which is then verified against the local TOTP secret.
  5. Upon successful verification, a temporary session is granted, allowing the agent to proceed with the protected task for a defined window.

feelgoodbot Setup

Ensure you have Go 1.21+ installed on macOS. Install the tool via go install:

go install github.com/kris-hansen/feelgoodbot/cmd/feelgoodbot@latest

Initialize the baseline and start the monitoring daemon:

feelgoodbot init
feelgoodbot daemon install
feelgoodbot daemon start

To configure step-up authentication for Openclaw Skills, initialize the TOTP secret:

feelgoodbot totp init --account "user@feelgoodbot"

feelgoodbot Data Schema & Taxonomy

The skill manages its state and configuration within the ~/.config/feelgoodbot/ directory using the following structure:

File/Folder Purpose
config.yaml Primary configuration for scan intervals and webhook alerts
totp.json Stores the encrypted TOTP secret and recovery backup codes
stepup-config.json Contains the registry of protected action patterns and patterns
snapshots/ Stores the file integrity baselines used for tampering detection
daemon.log Records monitoring activity, security events, and error logs

feelgoodbot Advanced Features

  • Pattern-based action protection supporting wildcards like payment:* or ssh:* for granular control.
  • Seamless integration with Clawdbot webhooks for real-time security alerting and notifications.
  • Automatic session caching for a 15-minute window to maintain a balance between high security and developer productivity.
  • Comprehensive recovery options including backup code generation and manual TOTP resets.
  • Built-in support for Openclaw Skills protocols to ensure multi-agent security consistency.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*