Skip to content

README.md

Latest commit

 

History

History
42 lines (33 loc) · 2.11 KB

README.md

File metadata and controls

42 lines (33 loc) · 2.11 KB

Explain Clawdbot (Gemini Analysis)

This directory contains a breakdown of the Clawdbot framework, generated by Gemini agents for a beginner audience.

Table of Contents

  1. What is Clawdbot?

    • A plain English explanation of the project.
    • Why it matters and who it is for.
    • The "Lobster" analogy.

  2. How It Works (Technical Overview)

    • The Architecture: Gateway, Channels, Agents.
    • Data flow diagrams.
    • Key concepts: Sandboxing and Plugins.

  3. Installation & Setup Guide

    • Scenario A: Standalone Mac Mini (Home Assistant).
    • Scenario B: Isolated VPS (Secure Cloud Bot).
    • Step-by-step instructions for both.

  4. Privacy & Security

    • Data locality and ownership.
    • The "Pairing" system for access control.
    • Docker sandboxing for safe code execution.

  5. Configuration Reference

    • clawdbot.json structure.
    • Critical security settings.
    • Environment variable overrides.

  6. Security Audit Report (2026)

    • Analysis of 512 security findings by Argus Security.
    • Key risks: Plaintext token storage, webhook signature bypass, and race conditions.
    • Link to full remediation plan and active discussion.

  7. Independent Security Audit (Saad Khalid)

    • Critical RCE: "Logic Bomb" allows agents to rewrite Docker startup commands via config.patch.
    • Auth Bypass: Agents can self-approve dangerous commands (missing role check in exec.approval.resolve).
    • Data Risks: Arbitrary file overwrite (nodes:screen_record) and arbitrary read (logs.tail).
    • Protocol Flaws: Token manipulation via pipe injection and environment variable hijacking (LD_PRELOAD).
    • Note: The reported DNS Rebinding (SSRF) appears mitigated in the current version via pinned dispatchers.