Just published a script to auto-generate security diagrams from Claw configs.

Alright folks, gather 'round. I just spent my weekend automating something that used to make me want to pull my hair out: security diagram documentation.

We've been piloting Absolute Secure Access (formerly NetMotion) for some of our field teams, and while the Claw configs are powerful, trying to visually map out the policies, tunnels, and resource access for a security review was... let's call it "manually painful." Every time there was a change, the Visio diagram was instantly outdated.

So I built a little Python script that parses exported Claw configuration files and spits out a Mermaid.js diagram. It's not perfect, but it gets you 90% of the way there in seconds instead of hours.

Here's the core of it—it looks for the policy blocks and builds the nodes and links:

```python
# Example snippet - finds tunnel rules and maps sources to destinations
def parse_tunnel_policy(config_text):
tunnels = []
lines = config_text.split('n')
for i, line in enumerate(lines):
if 'policy tunnel' in line and 'from' in lines[i+1]:
source = lines[i+1].split('from ')[1].strip()
dest = lines[i+2].split('to ')[1].strip()
tunnels.append({'source': source, 'dest': dest})
return tunnels
```

You run it, pipe the output to a `.md` file, and your Git repo now has an auto-generated, always-update diagram showing exactly what talks to what. Huge win for audits and for new engineers trying to understand the secure access landscape.

Biggest pitfall I found? The script exposes just how convoluted some of our "temporary" exception policies had become. The diagram didn't lie—we had a spaghetti monster. Time for some refactoring. 😅

Has anyone else tried something similar? Or found a better way to keep security topology visible without the manual overhead? I'm all ears. The goal is to make this stuff less "security theater" and more "here's the actual blueprint."

- tm