Skip to content

Forum

AI Assistant
Forums
Community
Show and Tell
 
Notifications
Clear all

Show and Tell

Original work worth sharing — hardening configs, threat model writeups, tooling you built, fuzzing results, or anything you learned the hard way. Posts should show your work, not just your conclusions.
RSS
Page 2 / 2

Step-by-step: Adding a mandatory human approval step for specific tool categories.

Last post by Markus Braun, 6 days ago

Posts: 6
Views: 4


Breaking: New CVE for a dependency Claw uses. Patching guide inside.

Last post by Emma Watson, 6 days ago

Posts: 18
Views: 6


Anyone else think the default system prompt is too powerful and needs to be constrained?

Last post by curious_leo, 6 days ago

Posts: 5
Views: 2


ELI5: Why can't I just run the whole thing in Docker and call it a day?

Last post by Tomislav Horvat, 6 days ago

Posts: 24
Views: 7


Thoughts on using WebAssembly for tool isolation instead of subprocesses?

Last post by Ray Chen, 6 days ago

Posts: 1
Views: 4


What is the best way to handle model file integrity? Checksums at load time?

Last post by Pete Nelson, 6 days ago

Posts: 11
Views: 3


Just released a set of OPA/Rego policies for validating agent action requests.

Last post by Jamie Rivera, 1 week ago

Posts: 13
Views: 1


Anyone else having issues with the memory isolation after upgrading to the latest dev branch?

Last post by Alex T., 1 week ago

Posts: 8
Views: 6


Page 2 / 2
Share:
Share
Tweet
Share