Total Visibility
Every file read, every API call, every command executed. Complete audit trail of what your AI agent does.
See Everything Your AI Agent Does
The observability layer for AI agents. Works with Claude Code, Docker Sandboxes, or standalone. Track every file access, API call, and command across Linux, Windows, and macOS. Real-time security alerts, cost tracking, and complete audit trails.
Works with Claude Code & Docker • Open source (MIT) • Free CLI
SudoDog is the observability layer for AI agents. Use it with Claude Code, Docker Sandboxes, or standalone.
See every action your agent takesβno matter how it runs.
Add SudoDog as a hook to Claude Code. Observe everything Claude does without changing your workflow.
sudodog integrate claude-code
Setup Guide
Run with Docker Sandboxes and add SudoDog for deep observability inside the container.
sudodog integrate docker
Learn More
Run any agent with SudoDog CLI. Works with LangChain, AutoGPT, CrewAI, or custom agents.
sudodog run python agent.py
Install CLI
Zero integration - just prepend sudodog run to your existing command
user@laptop:~$ sudodog run python agent.py
π SudoDog AI Agent Security
ββββββββββββββββββββββββββββββββββββββββ
π Platform: macOS
βΆοΈ Running: python agent.py
[Safe Operation] Reading configuration...
[Safe Operation] Connecting to API...
[Security Alert] Attempting SQL injection pattern detected
Pattern: DROP TABLE users;
β οΈ Blocked by SudoDog
β
Command completed successfully
CPU: 8.2% | Memory: 124MB
user@laptop:~$
Every file read, every API call, every command executed. Complete audit trail of what your AI agent does.
Claude Code, Docker Sandboxes, or standalone. SudoDog provides observability regardless of how you run your agent.
Watch your agents in action. Live CPU, memory, and network stats. Know exactly what's happening right now.
Detect suspicious patterns in real-time. SQL injection attempts, sensitive file access, unauthorized API calls.
Know exactly which agent action cost you money. Track LLM spend by task, by agent, by team.
Immutable audit logs for SOC 2, EU AI Act, GDPR. Export reports with one click.
Monitor locally for free. Upgrade to dashboard when you need team visibility.
$0/month
Monitor agents locally on any platform
Free Beta
Web dashboard for teams
Free Beta
Advanced compliance and security
Install on your platform and start monitoring immediately. No code changes required.
# Install from PyPI (Linux, macOS, Windows)
pip install sudodog
# Scan for AI agents running on your machine
sudodog-scan
# Run your agent with monitoring
sudodog run python agent.py
# Or integrate with Claude Code
sudodog integrate claude-code
# Linux
curl -sL https://sudodog.com/install-linux.sh | bash
# macOS
curl -sL https://sudodog.com/install-macos.sh | bash
# Windows (PowerShell as Administrator)
irm https://sudodog.com/install.ps1 | iex
Zero code changes - just prepend sudodog run
# LangChain
sudodog run python langchain_agent.py
# AutoGPT
sudodog run python -m autogpt
# CrewAI
sudodog run python crewai_agent.py
# Custom agent
sudodog run python your_agent.py
# Any language
sudodog run node agent.js
sudodog run ruby agent.rb
π Full Documentation: Documentation β’ Quick Start Guide
Open Source: View on GitHub β’ MIT License