Three Open-Source Tools for Monitoring Your Connections, Prompts and Agents

Most developers can’t see (or don’t often look at) the traffic happening on their machines, or what kinds of data their computer or their AI tools send out.

These four open-source tools are great ways for you to monitor (or block) network traffic as well as your AI usage, to see exactly what’s going on under the bonnet.

1. Snitch: See every connection

Snitch shows every connection your computer makes, live, as a graph and world map, and which app is behind each one.

It is worth noting Snitch isn’t a blocker, so we recommend paring it with a firewall, but it still gives you a good picture of what’s happening in the background.

Check out the Snitch repo on GitHub here.

2. OpenSnitch: Block connections

OpenSnitch is a Linux firewall that works per app. Whenever a program tries to connect out, it asks you to allow once, allow always or block. Linux’s built-in firewall tools only filter by address and port, so they can’t tell you which app is calling out. Expect a noisy first week of pop-ups while you vet your apps.

See more over on the OpenSnitch GitHub repo.

3. moatleak: Scanning logs

moatleak is a brand new project from AndyJBryant that reads the logs that your AI tools already keep and reports which companies, and which countries, handled your prompts.

After installing, just run moatleak scan to get your results, though it’s worth saying moatleak only reads Claude Code and Hermes logs, for now but why not open a pull request if you think you can improve it.

Check it out over on the moatleak GitHub repo.

A simple way to think of these tools and how to use them is that Snitch shows the connections your machine is making, OpenSnitch gives you the ability to get ahead of these connections to block them before they occur, and moatleak offers a similar kind of overview but for LLMs.

That’s it for now, be sure to let us know what you think on X, Instagram, YouTube or LinkedIn and until next time… happy building!